INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
Unraveling The Nexus: Capital Flows Dynamics, Financial Sector  
Stability and Their Impact on Economic Development in Nigeria  
Oluwatosin Yewande Akinbode (Ph.D), Ofonime Moses Akpan (Ph.D)  
Department of Economics University of Uyo  
Department of Economics School of Arts and Social Sciences College of Education, Afaha Nsit  
DOI: https://dx.doi.org/10.47772/IJRISS.2025.91100286  
Received: 10 November 2025; Accepted: 20 November 2025; Published: 06 December 2025  
ABSTRACT  
Capital flows Dynamics, Financial sector stability and their impact on Economic development in Nigeria was  
carried out using data from 1986 to 2022. This study attempted to find out the effect of the relationship  
between financial sector stability variables and capital flows on economic development in Nigeria. The ARDL  
Error Correction regression analysis was used to test the short run and long run relationship capital outflow  
and financial sector stability on economic development. The study found out that capital flow variables (NPI  
and NDI) negatively but significantly affect economic development while financial sector variables (CPS and  
NIM) exhibited a negative and insignificant relationship with economic development. The study concluded  
that the influence of capital flows is dependent upon a nation’s degree of financial development. A more open  
capital account has a detrimental impact on performance for nations with low levels of financial development.  
The study therefore recommends that the policy makers in Nigeria should strike a balance between attracting  
foreign investment and ensuring that it aligns with broader economic development goals and ensure that credit  
is directed towards productive sectors that contribute to economic development.  
INTRODUCTION  
The pursuits of economic development, exchange rate stability, low inflation, financial sector stability and  
sustainable balance of payment (BOP) have over time been the force behind most economic policies. The  
realization of these laudable objectives has no doubt been constrained by the interplay of factors, among,  
which include low level of domestic savings and investment and foreign exchange shortage. The emergence of  
integrated financial markets and high capital mobility made possible by the increasing globalization of world  
economies, has predisposed economies, especially developing ones to the volatility of capital flows,  
dependence on financial sector for capital and loss of market confidence, which often result in severe financial  
crises.  
Capital flows dynamics refer to the movement of funds into and out of a country, which can be in the form of  
foreign direct investment (FDI), portfolio investments, or remittances. These capital flows play a significant  
role in shaping the financial sector's stability and overall economic development in Nigeria.  
The stability of the financial sector is vital for fostering economic growth as it ensures efficient allocation of  
resources, promotes financial inclusion, and reduces systemic risks. A well-functioning financial sector can  
facilitate capital mobilization, channel funds to productive sectors, and support entrepreneurship and  
innovation.  
However, capital flows dynamics can also pose challenges to financial stability. Sudden capital outflows, for  
example, can lead to currency depreciation, inflation, and financial market volatility. Improperly managed  
capital flows can also increase vulnerability to external shocks and undermine economic growth prospects.  
Most developing countries are characterized by low level of domestic savings, which has impeded the much-  
needed investment for economic development. In order to attain a desirable level of investment that would  
Page 3650  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
ensure sustainable development, developing country needs some foreign savings to bridge the savings-  
investment gap. The gap when financed through foreign savings comes in form of capital flows. Capital flows  
is transmitted through foreign direct reserves, foreign loans and credits etc (Obadan, 2004).  
Capital flows in terms of portfolio investment has been a notable feature of developed economies. This,  
however, is becoming a very important component of the balance of payments of many emerging economies,  
such as China, Hong Kong, India, Singapore, Taiwan, Brazil, South Africa etc (Obadan, 2004).  
In Nigeria, the abrogation of certain laws and subsequent entrenchment of investment friendly laws as well as  
the introduction of structural reforms facilitated the substantial flow of capital. Until 1986, Nigeria did not  
record any figure on portfolio investment (inflow or outflow) in her BOP accounts. This was attributable to the  
non-internationalization of the country ‘s money and capital markets as well as the non-disclosure of  
information on the portfolio investments of Nigerian investors in foreign capital/money markets (CBN  
1997:151). For example, the net portfolio investment (NPI) and net direct investment (NDI) were N151.6  
million and N735.8 million in 1986, which rose to N51, 079.13 million and N115, 952.2 million in 2000,  
indicating a growth rate of 33,593.36 and 15,658.66 per cent, respectively. In 2005, NPI and NDI went up to  
N116, 035.00 million and N654, 193.10 million indicating a growth rate of 127.17 and 464.19 per cent,  
respectively, compared with the 2000 figures. Furthermore, NPI and NDI outflow supersedes inflow to  
N560498.52million and N124645.02 million in 2008, respectively, this figure further decreases to about  
329409million and 84768.5million in 2015. Also, in 2022 NPI decline to 294350million while NDI increase to  
28535.65million.CBN (2022)  
In recent years, Nigeria has experienced significant fluctuations in capital flows, accompanied by varying  
degrees of instability in its financial sector. This phenomenon has prompted a growing interest in  
understanding the intricate relationship between capital flows dynamics, financial sector stability, and their  
implications for economic development in the country. This research endeavors to delve into this complex  
nexus to shed light on the underlying mechanisms and potential policy implications.  
Despite Nigeria's status as one of Africa's largest economies, it has grappled with persistent challenges related  
to capital flows volatility and financial sector fragility. The erratic nature of capital inflows and outflows,  
coupled with vulnerabilities within the financial system, has posed considerable hurdles to sustainable  
economic growth and development. Moreover, the interplay between these factors remains poorly understood,  
hindering the formulation of effective policy measures to mitigate risks and harness opportunities for economic  
advancement.  
The study will provide valuable insights into the drivers and determinants of capital movements, enabling  
policymakers to adopt proactive measures to manage volatility and enhance resilience. It will create a better  
understanding of how fluctuations in capital flows impact financial sector stability is crucial for safeguarding  
the integrity of the banking system and promoting investor confidence  
Moreover, this research holds significance not only for Nigeria but also for broader debates surrounding  
emerging market economies and their quest for sustained growth and resilience in an interconnected world. By  
shedding light on the nuanced interrelationships between capital flows, financial sector stability, and economic  
development, it seeks to inform policymakers, practitioners, and scholars alike, offering insights that can guide  
policy formulation, risk management strategies, and institutional reforms. In doing so, it aspires to contribute  
to the advancement of knowledge and the pursuit of inclusive and sustainable development in Nigeria and  
beyond at fostering sustainable and inclusive growth in Nigeria. By addressing these issues, this research seeks  
to contribute to the ongoing discourse on economic policy formulation and implementation in the country, with  
the ultimate goal of fostering a more robust and resilient economy.  
The paper is structured into 5 sections. Following the introduction is section 2, which reviews the theoretical  
literature. In section 3 is the methodology. Section 4, deals with the analysis. Finally, section 5 discusses the  
summary, conclusion as well as recommendations.  
Page 3651  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
LITERATURE REVIEW  
This section reviews the relevant literature on capital flows dynamics, financial sector stability, and economic  
development. To enhance clarity, the review is divided into three subsections: theoretical review, empirical  
review, and contextual review. The theoretical review discusses foundational theories underpinning the study.  
The empirical review examines global studies on the relationships among the variables. The contextual review  
focuses on Nigeria-specific studies and evidence to situate the research within the local economic landscape.  
THEORETICAL REVIEW  
Two theories underpinned this study they are the Internalization theory and the integrated financial  
liberalization and financial repression theory seeks Internalization theory, initially proposed by Stephen Hymer  
in the 1960s and further developed by other scholars such as John Dunning, addresses why firms engage in  
foreign direct investment (FDI) instead of merely conducting international trade or licensing agreements. The  
theory suggests that firms undertake FDI to internalize market imperfections and gain control over foreign  
assets, thereby maximizing their profits. Internalization theory starts with the recognition of market  
imperfections, such as imperfect information, transaction costs, and incomplete contracts. In a perfect market,  
firms would have no incentive to invest abroad; they could simply trade or license their products or  
technologies. However, in the real world, these market imperfections exist and create inefficiencies that firms  
seek to overcome.  
The theory suggests that firms internalize foreign operations to capture the benefits of coordination, control,  
and coordination economies. By owning and managing foreign subsidiaries, firms can coordinate production,  
marketing, and distribution activities more efficiently, reducing transaction costs and enhancing overall  
profitability.  
John Dunning extended Hymer's ideas and proposed the OLI framework, which combines three factors -  
Ownership advantages (O), Location advantages (L), and Internalization advantages (I). According to  
Dunning, firms engage in FDI when they possess ownership advantages, operate in locations with attractive  
investment opportunities (such as favorable market conditions, resources, or infrastructure), and can internalize  
the benefits of control and coordination.  
In summary, internalization theory provides a compelling explanation for why firms engage in FDI by  
highlighting the role of market imperfections and the advantages of internalizing foreign operations. By  
gaining control over foreign assets and activities, firms can exploit their ownership advantages more  
effectively, reduce transaction costs, and enhance their competitiveness in the global marketplace.  
The Integrated financial liberalization and financial repression theory was propounded by economists such as  
Joseph Stiglitz, Dani Rodrik, and Ha-Joon Chang. They argue that a balanced approach to financial sector  
development, which combines market discipline with government oversight, is essential for achieving  
sustainable and inclusive economic growth. By harnessing the benefits of financial liberalization while  
mitigating its potential downsides, countries can create a more resilient and equitable financial system that  
supports long-term development objectives.  
The other theory, the integrated financial liberalization and financial repression theory seeks to reconcile the  
potential benefits and risks associated with financial liberalization and government intervention in the financial  
sector. The theory recognizes that both approaches have their advantages and disadvantages, and proposes a  
balanced approach that combines elements of liberalization with regulatory measures to promote economic  
development while mitigating financial instability.  
Proponents of this theory argue that financial liberalization can stimulate economic growth by fostering  
competition, innovation, and efficiency in financial markets. By removing restrictions on capital flows, interest  
rates, and financial intermediation, liberalization allows for greater access to credit, encourages investment,  
and facilitates the allocation of resources to their most productive uses. This, in turn, can lead to higher levels  
of investment, employment, and overall economic development.  
Page 3652  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
However, proponents also acknowledge the potential risks associated with unchecked financial liberalization,  
including increased volatility, asset bubbles, and systemic instability. Financial markets may become more  
prone to speculative excesses, leading to boom-bust cycles and financial crises. Moreover, liberalization can  
exacerbate income inequality and exacerbate social tensions if the benefits accrue disproportionately to a small  
segment of the population.  
To address these concerns, proponents of integrated financial liberalization advocate for a regulatory  
framework that combines market-oriented policies with prudential measures to ensure financial stability and  
promote inclusive growth. Regulatory interventions may include capital adequacy requirements, liquidity  
standards, risk management guidelines, and consumer protection measures. Additionally, macroprudential  
tools such as reserve requirements, loan-to-value ratios, and countercyclical buffers can help mitigate systemic  
risks and dampen excessive credit expansion during economic booms.  
Empirical review  
Capital inflows may lead to improvement in the financial sector to reduce vulnerability to crisis, and have a  
positive impact on macroeconomic stability in financially open economies (Kose et al. 2010). As the financial  
sector becomes more developed, the growth benefits of capital flows will improve. Bekaert and Harvey (2000),  
for instance, found that with increased capital flows, risk is reduced, equity returns become highly correlated  
with the world market, per capita GDP increases marginally, and inflation and foreign exchange rate volatility  
are lowered.  
Macroeconomic stability has implications for the volume and composition of capital flows, as developed  
financial markets moderate the effects of shocks and helps reduce macroeconomic volatility (Kose et al. 2010).  
Macroeconomic implications of financial globalization are experienced through the effects on economic  
growth and growth volatility (Kose et al. 2010). Thus, capital inflows enhance growth more in countries with  
strong macroeconomic policies and where there is macroeconomic stability (Eichengreen 2000). Policy also  
plays a significant role in explaining changes in the level of inflows and their volatility. Alfaro, Kalemli-  
Ozcan, and Volosovych 2007.  
Variations in capital flows to developing countries are mostly explained by shocks to real variables of  
economic activity, such as foreign output and domestic productivity. De Vita and Kyaw (2008). The impact of  
capital flows also depends on the level of financial development of a country. Countries at low levels of  
financial development experience a negative effect on performance, in the case of a more open capital account  
Edwards (2001). Choong et al. (2010) found private capital flows to positively impact growth in countries with  
well-developed financial sectors but have negative effects in situations of poor financial sector development.  
The level of capital flows, however, depends on the degree of market integration, which is measured by  
differences in rates of return across countries (Frankel 1992). As Mohan and Kapur (2010) found, the  
increasing volume of private capital flows to emerging market economies depends on, among other factors,  
their growing degree of financial openness over time, growth in overall profitability of firms, positive interest  
differentials in favor of these economies, and the expectation of continuing currency appreciation.  
The link between capital flows, financial sector stability and economic development has been studied by  
testing their causal relationship individually. Studies on the relationship between capital flows and growth, and  
financial stability and growth however, are inconclusive in their findings.  
Harley T.W and Akinola A. T (2018) studied the impact of capital flows on the financial system stability of the  
Nigeria economy using data from 1987 to 2017. The study found out that credit to private sector negatively  
affect foreign direct investment. The coefficient of determination in the study was high indicating that the  
explanatory variables (financial system stability variables) are captured by the capital flows variable (FDI).  
The study therefore recommended that the Nigeria government should adjust the model for credit to private  
sector so as to have a positive relationship with capital flows.  
Justine C, (2018). Examined whether sharp capital flow movements, specifically private capital inflow  
movements, are a significant risk factor for financial stability in Jamaica. A structural vector autoregressive  
Page 3653  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
(SVAR) model was used to assess the dynamic relationship between private capital inflows and financial  
stability, as well as the responsiveness of financial stability indicators to sudden changes in private capital  
flows. The findings confirmed a significant relationship between private capital inflows and financial stability  
and underscores the need to develop macro-prudential measures to curb possible threats to financial stability.  
Nyang`oro, O. (2017). Analyzed the effect of capital flows on economic growth in sub-Saharan Africa, using  
a system of generalized methods of moment (GMM) model. It tests the extent to which the level and volatility  
of capital inflows, both disaggregated and total, affect economic growth. The study finds that portfolio equity  
has a positive effect on economic growth while private equity and debt are inversely related to growth.  
However, volatility of portfolio equity and private equity has no impact on economic growth, pointing to low  
levels of financial integration in these countries. Total capital inflows, both gross and net inflows, have a  
negative effect on growth, while volatility of total gross capital inflows has a positive effect, and that of total  
net capital inflows is positively related to growth. The effect of total capital inflows is possibly influenced by  
the overall effect of debt in these economies. The findings suggested that concerns on capital inflows should  
mainly be addressed through the debt market, and that the growth benefits of capital inflows can be achieved  
by improving financial markets, ensuring macroeconomic stability, and having in place good institutions.  
Using cointegration and panel Granger causality tests, Abbes et al. (2015) found economic growth and FDI to  
be cointegrated in the long run, in a sample of 65 countries from 1980–2010. However, they established a  
unidirectional causality from FDI to GDP. Albulescu (2015) found both direct and portfolio investments  
exerted an influence on long-term economic growth; when equity and investment funds instruments were  
considered, Aizenman, Jinjarak, and Park (2011) found a positive effect of FDI on growth but no effect from  
portfolio inflows and equity investment, while short-term debt has no effect before a crisis period and a  
negative effect during the crisis.  
Contextual review  
However, few studies have examined the capital flow dynamics and financial sector stability on economic  
development in Nigeria. The literature points to the various macroeconomic variables that are affected by  
capital Inflows and financial stability. The effects are not standard across countries, as they depend mainly on  
the type of capital flowing to a country, the level of financial development, and the macroeconomic state of a  
country.  
Despite empirical studies on the link between capital flows, financial sector stability and economic  
development, limited attempts have been made to connect financial sector stability and volatility of capital  
flows. This study establishes the individual relationship between the level and volatility of capital inflows and  
economic development; financial sector stability and economic development. Capital inflows, measured using  
net foreign direct investment and net portfolio investment while financial sector stability is measured using  
credit to private sector and net interest margin to test their differential effects on macroeconomic variables.  
METHODOLOGY  
This study seeks to determine the impact of capital flow dynamics and financial sector stability on economic  
development in Nigeria This section discusses the research design, model specification, variables used, sources  
of data, and finally, estimation technique and procedure used for the study.  
Research design  
Given the nature of this research, the study made use of annual time series data to determine the relationship  
between capital flow and financial sector variables on economic development in Nigeria. The study employed  
the ARDL Bounds test analytical technique. The secondary data covering the period of 1986 to 2022 will be  
used.  
Page 3654  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
Model Specification  
The main objective of this study is the impact of capital flows on financial system stability of the Nigeria  
economy. It applies the ARDL Bounds test analytical technique based on the traditional determinants of  
financial stability and capital flows distilled from the literature. The idea is to subject the variables to a linear  
model and test the impacts of financial stability variables and capital flows on economic development. We  
describe the indicators used, data sources, and the estimation technique applied in the empirical investigation  
of the relationship between capital flows and financial stability. For the time period 1986–2022, a set of  
financial indicators was used to measure financial system stability, gross capital flows, and a macroeconomic  
variable as control variable.  
In order to account for the impact of capital flows on financial system stability of the Nigeria economy, the  
model for the study is hereby specified as follows:  
Model 1: capital flow and economic development  
GDP= a0 + α1NPI+ α2 NDI + α3 EXRATE + α4 INF + εt …………… 1  
Model 2: financial sector and economic development  
GDP = a0 + α1CPS+ α2 NIM + α3 EXRATE + α4 INF + εt …………… 2  
Where:  
GDP = gross domestic product per capita (proxy for economic development)  
NPI = Net port-folio investment (proxy for capital flow)  
NDI = Net Direct investment (proxy for capital flow)  
CPS = credit to private sector  
NIM = net interest margin  
EXRATE = exchange rate  
INF = inflation rate  
Apriori Expectation  
Foreign port-folio investment and foreign direct investment are proxy for capital flows and it is expected that  
FPI and FDI be should be positive  
Net interest margin is expected that increase in interest margin should lead to increase in financial system  
stability  
Credit to private sector is expected that an increase in credit to the private sector should lead to increase in the  
Nigeria financial system stability  
Inflation rate is expected that high inflation rate will impede the Nigeria financial system stability.  
Exchange rate fluctuations can influence foreign investment flows. A weaker currency may attract foreign  
investment which can boost domestic income. Conversely, a strong currency might discourage foreign  
investment or even lead to capital flight.  
Page 3655  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
Description of Variables  
GDP per capita is selected as a proxy for economic development consistent with global development literature.  
Net Portfolio Investment (NPI) and Net Direct Investment (NDI) measure capital flow dynamics as reported by  
the CBN. Credit to Private Sector (CPS) and Net Interest Margin (NIM) capture financial sector stability,  
following indicators recommended by the IMF and World Bank. Inflation and exchange rate are included as  
control variables commonly affecting macroeconomic stability (Kose et al., 2010).  
Estimation technique  
Unit root test  
In time series analysis, before running the co-integration test the variables must be tested for stationarity. For  
this purpose, we use the conventional ADF tests to test for stationarity.  
Co-integration Approach  
This study relies on the ARDL bounds test approach to co-integration developed by Pesaran et al., (2001) to  
test for cointegration. The ARDL bounds test approach to co-integration has been demonstrated to perform  
better than other traditional cointegration methods. This is because of its numerous advantages over other long  
run estimation techniques. It can be applied on variables that are either I(1) or I(0) or combination of the two  
and the approach yields unbiased estimates and its t-statistics are effective even if some of the regressors are  
endogenous (Harris & Sollis 2003).  
Model Justification  
The ARDL approach is adopted due to its flexibility in handling variables integrated at both I(0) and I(1), its  
efficiency in small sample sizes, and its ability to provide both long-run and short-run estimates (Pesaran et al.,  
2001). ARDL also mitigates endogeneity concerns that frequently appear in macroeconomic datasets.  
Thus, we specified conditional general form of the ARDL model in equation 3 and 4  
∆GDPt=α0+α1GDPt−1+α2NPIt−1+α3NDIt−1+α4EXRATEt−1+α5INFt−1+  
∑
∑
∑
∑
∑
θ ∆GDPt−1+  
θ ∆NPIt−1+  
θ ∆NDIt−1+  
θ EXRATEt-1+  
θ INFt-1+µt  
1
2
3
4
3
=
=
=
=
=
……………………………………. (3)  
∑
∑
∆GDPt=α0+α1GDPt−1+α2NIMt−1+α3CPSt−1+α4EXRATEt−1+α5INFt−1+  
θ ∆GDPt−1+  
θ ∆NIMt−1+  
1
2
=
=
∑
∑
∑
θ EXRATEt-1+  
θ ∆CPSt−1+  
θ INFt-1+µt ………………………. (4)  
3
4
3
=
=
=
Where: GDP, NPI, NDI, NIM, CPS, EXRATE, and INF remain as previously defined. Similarly, α denotes the  
drift, v denotes the lag lengths, α1 – α5 are coefficients to be estimated while ln denotes natural logarithms and  
µt is the stochastic error term  
Since the aim of the study is to understand both short and long run dynamics of capital flows, financial sector  
impact on economic development in Nigeria, specification of the long run and short run ARDL approach is  
important. Hence, the long run model is expressed in equation 5 and 6 below:  
∑
∑
∑
∑
∑
α4EXRATEt−1+  
GDPt=α0+ =1 α1GDPt−1+  
α2NPIt−1+  
α3NDIt−1+  
α5INFt−1+  
=
=
=
=
µt……………………………………… (5)  
α
∑
∑
∑
∑
∑
α4EXRATEt−1+  
GDPt=α0+ =1 α1GDPt−1+  
α2NIMt−1+  
CPSt−1+  
α5INFt−1+  
=
=
=
=
3
µt……………………………………… (6)  
Similarly, to estimate the short run parameters of the model when the long run equilibrium exist, the  
unrestricted ARDL of error correction model is estimated as captured in equation (7 and 8):  
Page 3656  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
∑
∑
∑
∑
∆GDPt=α0+ =1 α1 ∆GDPt−1+ =1 α2 ∆NPIt−1+ =1 α3 ∆NDIt−1+ =1 α4  
∑
+
∆EXRATEt−1  
=1 α5 ∆INFt−1 +θECMt−i +µt ………………….. (7)  
∑
∑
∑
∑
∆GDPt=α0+ =1 α1 ∆GDPt−1+ =1 α2 ∆NIMt−1+ =1 α3 ∆CPSt−1+ =1 α4  
∑
+
∆EXRATEt−1  
=1 α5 ∆INFt−1 +θECMt−i +µt ………………….. (8)  
Where α1, α2, α3, α4 and α5 are short-run coefficients of the ARDL model, α0 is the constant, θ is the speed of  
adjustment in the system and ECM denotes the stochastic error term.  
Diagnostic Tests  
Diagnostic tests were applied to validate model reliability.  
- Serial correlation was assessed using the Breusch–Godfrey LM test.  
- Heteroskedasticity was evaluated using White’s test.  
- Normality was confirmed using the Jarque–Bera statistic.  
- Model stability was verified using CUSUM and CUSUMSQ tests, following Brown et al. (1975).  
Analysis  
This section presents the results of the various estimation techniques used to achieve the objective of the study.  
Statistical Data Analysis  
The graphical analysis of the time series of variables is presented in this section.  
Graphical Presentation of Data  
Figure 1: Net Port-folio investment, Net Direct Investment and GDP per capita from 1986-2022  
1,200,000  
800,000  
400,000  
0
-400,000  
-800,000  
1990  
1995  
2000  
NPI  
2005  
NDI  
2010  
GDP  
2015  
2020  
Capital Flows, Financial Market and Economic Growth in Nigeria  
The introduction of Structural Adjustment Programme (SAP) in 1986 marked an epoch in the liberalization of  
the Nigerian economy. Prior to the period, the economy was predominantly regulated, that affected the free  
movement of capital necessary for economic growth. SAP heralded a lot of policy reforms that led to the  
publication of an Industrial Policy for Nigeria in January 1989. Critical policy reforms leading to the changes  
in the investment climate in Nigeria for both domestic and foreign investors (provision of enormous  
Page 3657  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
opportunity to participate in the economy) were the abrogation of the Nigerian Enterprises Promotion Decree  
1989 and the Exchange Control Act of 1962 as well as their subsequent replacements with the Nigerian  
Investment Promotion Council Decree No 16 of 1995 and Foreign Exchange (Monitoring and Miscellaneous  
Provisions) Decree 17 of 1995. As mentioned earlier, the country did not record any NPI on her BOP until  
1986. Onosode (1997) posited that between July 1995 and July 1996, about US$6.0 million foreign portfolio  
investment (FPI) was made in the Nigerian capital market through the Nigerian Stock Exchange (NSE) for the  
first time since 1962, while for the whole of 1996, foreign investment through the Nigerian Stock Exchange  
totaled UD$32.99 million.  
In 1986, the NPI in Nigeria was N151.6 million. It rose to N51, 079.13 million in 2000. By 2005, there was a  
tremendous increase in the NPI figure in Nigeria. It increased from N51, 079.13 million to N116, 035.00  
million from 2000 to 2005, (a growth rate of 127.17 per cent). It marked the period when the banks were  
statutorily mandated to shore up their capital base from mere N2.0 billion to N25.0 billion. It rose to a record  
level of N231942 million in 2007 before declining to N122347 Million in 2009 and further decline to 309158  
and 294350 in 2021 and 2022 respectively. Similarly, the NDI was N735.8 million in 1986 and rose to N115,  
952.16 million in 2000. It further increased from N654, 193.10 million in 2005 to N1, 779,594.80 million in  
2006, indicating a growth rate of 172.02 per cent. There was a net outflows of direct investment to N109,  
161million and 124,645million in 2007 and 2008 respectively before rising to N28535.65 million in 2022.  
During this period, GDP per capita was N2809.015million in 1986. It rose to N57, 489.92million in 2000 and  
stood at N925, 981.1 in 2022. Comparatively, the NPI and NDI recorded an average annual figures of  
N108,658million and N50,567.8million during 1986 - 2022.  
Figure 2: credit to private sector and net interest margin  
24  
20  
16  
12  
8
4
0
1990  
1995  
2000  
2005  
2010  
2015  
2020  
CPS  
NIM  
Likewise, in figure 2 above there was evidence of fluctuation in the financial sector variables, as credit to  
private sector and net interest margin shows a period of increase and decrease. CPS was at its highest in 2009,  
as a result of some reforms in the financial sector, this period NIM was at its lowest. This was to in other to  
attract investment in the economy.  
Unit Root Tests  
Table 1, summarizes the results obtained for each variable from the various techniques used to test the  
hypothesis of unit root or no unit root as the case may be.  
Table 1: ADF Unit Root Tests for all Variables  
Variables  
Levels  
First  
Order of  
Integration  
I (1)  
Difference  
-4.158658  
-6.439144  
-4.296517  
-3.912883  
GDP  
NPI  
I (1)  
NDI  
CPS  
I (1)  
I (1)  
Page 3658  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
NIM  
-3.550393  
I (0)  
I (1)  
I (1)  
EXRATE  
INF  
-5.604783  
-4.296517  
Source: Authors’ computation using (EVIEWS 12)  
The Augmented Dickey Fuller (ADF) unit test results shows that all the variables are stationary at first  
difference except net interest margin (NIM) which is stationary at levels.  
Model Estimation  
Following the unit root which shows all variables are stationary at first different and levels, there is an  
econometric justification to apply the ARDL estimation technique. As such, the ARDL bond test and  
coefficient estimation was used for each of the specified model and the result is presented as follows;  
Model 1 on Capital outflow and economic development  
TABLE 2 ARDL Bond Test Result for Model 1  
Model  
F Statistics  
5% Critical  
value  
Decision  
Equation 1  
10.70  
Co- integration  
I(0)  
I(1)  
3.49  
2.56  
Source: Authors’ computation using (EVIEWS 12)  
The ARDL bounds test is based on the assumption that the variables are I (0) or I (1) as shown above in the  
unit root table. The results of the ARDL bounds testing approach are presented in table (2) indicating that the  
computed F-statistics for explanatory variables was (10.70). The f- bound test statistics of (10.70) exceeds  
upper critical bound (3.49) at 5% level of significance. This finding supports that co-integrating relationship  
exists and confirms the stable long-run relationship between the variables. This implies that the null hypothesis  
of no co-integration among the variables is rejected.  
Table 3: ARDL Error Correction (ECM) Regression for Model 1  
Case 2: Restricted Constant and No Trend  
Variable  
Coefficient  
Std. Error  
t-Statistic  
Prob.  
D(GDP(-1))  
D(GDP(-2))  
D(GDP(-3))  
D(NPI)  
D(NPI(-1))  
D(NPI(-2))  
D(NPI(-3))  
D(NDI)  
D(NDI(-1))  
D(NDI(-2))  
D(NDI(-3))  
D(EXRATE)  
D(EXRATE(-1))  
D(EXRATE(-2))  
D(EXRATE(-3))  
D(INF)  
D(INF(-1))  
D(INF(-2))  
D(INF(-3))  
CointEq(-1)*  
-0.013927  
-0.154846  
-0.841945  
-0.091612  
0.071996  
0.033161  
0.021692  
-0.016124  
0.048507  
0.039730  
0.133823  
55.55277  
-333.3141  
-330.7031  
-155.0810  
-199.0215  
-106.1070  
-274.0704  
-76.76570  
-0.247472  
0.102822  
0.125238  
0.127750  
0.010196  
0.008726  
0.009821  
0.009744  
0.008013  
0.015141  
0.014266  
0.017127  
44.54259  
69.30068  
61.57887  
51.42325  
62.28771  
62.78011  
58.78612  
52.94694  
0.024218  
-0.135448  
-1.236412  
-6.590578  
-8.985434  
8.251156  
3.376459  
2.226172  
-2.012099  
3.203804  
2.785007  
7.813651  
1.247183  
-4.809680  
-5.370399  
-3.015776  
-3.195197  
-1.690137  
-4.662161  
-1.449861  
-10.21838  
0.8956  
0.2514  
0.0002  
0.0000  
0.0000  
0.0097  
0.0566  
0.0790  
0.0125  
0.0237  
0.0001  
0.2476  
0.0013  
0.0007  
0.0167  
0.0127  
0.1295  
0.0016  
0.1851  
0.0000  
R-squared  
0.999920 Mean dependent var  
0.999682 S.D. dependent var  
4945.404 Akaike info criterion  
1.96E+08 Schwarz criterion  
-304.1484 Hannan-Quinn criter.  
4188.791 Durbin-Watson stat  
0.000000  
286690.9  
Adjusted R-squared  
S.E. of regression  
Sum squared resid  
Log likelihood  
F-statistic  
Prob(F-statistic)  
277200.5  
19.94839  
21.08211  
20.32985  
2.063749  
The result showed that R2 (R-Squared) value of 0.999 was obtained. This implies that 99% of the variations in  
economic development is explained by changes in net port-folio investment, net direct investment, exchange  
Page 3659  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
rate and inflation. The F-Statistics of 4188.791 is significant considering the probability value. This implies  
that the model has goodness of fit. The Durbin-Watson value of 2.06 which is approximately 2 indicates that  
there exists no serial correlation. The error correction model (ECM) for economic development is specified,  
following the Representation Theorem specified. The primary reason for this is to capture the dynamics in  
economic development and capital flow in the short-run and identify the speed of adjustment as a response to  
departures from the long-run equilibrium. Thus, the general specification framework of the ECM is shown in  
Table 3.  
Statistically, the ECM term is negative and significant at 5% probability level. The existence of short run  
equilibrium among the time series in economic development and capita flow is validated by this result. The  
slope coefficient of the error term in absolute terms (0.247) represents the speed of adjustment and is  
consistent with the hypothesis of convergence towards the long-run equilibrium once economic development  
and capital flow equation fluctuates from its equilibrium in the short run. The coefficient of the ECMt-1  
suggests that economic development adjusts to the explanatory time series as about 25 percent of the  
discrepancy between long-run and short-run is corrected annually in the country, which is very low.  
The coefficient of net port-folio investment (-0.091612) exert a negative but significant relationship with  
economic development (GDP) in the current period. Since, NPI (Net Private Investment) is negative and  
statistically significant to GDP, as this does not conform to a priori expectation, it suggests that decreases in  
private investment have a significant negative impact on GDP. The policy implication of this finding would be  
for policymakers to focus on measures to stimulate private investment in order to support GDP growth. Such  
as; Investment incentives, Infrastructure spending, Access to finance and Economic stability measures.  
However, NPI lagged one and two with coefficient (0.071996 and 0.033161) respectively are both positive and  
statistically significant to GDP that is a 1% increase in net portfolio investment will lead to 7.2% and 3.3%  
increase in GDP. It suggests that past increases in net private investment have had a significant positive impact  
on current GDP. The policy implication of this finding is that policies supporting and encouraging private  
investment such as incentives, subsides and infrastructural development could be effective in stimulating  
economic development.  
Similarly, Net direct investment with coefficient (-0.016124) has a negative and insignificant relationship with  
GDP. This implies that current levels of net direct investment do not have a statistically significant impact on  
GDP. However, the coefficient of NDI at lag 1, 2 and 3 has a positive and significant impact with GDP, which  
suggests that past levels of net direct investment have a significant positive effect on GDP. The policymakers  
should encourage long-term investment, promote stability and address barriers to investment to support  
sustained economic development.  
The coefficient of exchange rate shows a positive and insignificant relationship in the current period to GDP,  
while lagged exchange rates at one, two, and three periods are positive and statistically significant to GDP, it  
suggests that past changes in the exchange rate have a significant impact on current GDP, but the current  
exchange rate alone does not influence GDP significantly. Policymakers may need to focus on maintaining  
exchange rate stability over the long term to support economic development. Volatility in exchange rates could  
disrupt economic activity and hinder growth, especially as past fluctuations have had significant impacts on  
GDP.  
Finally, the coefficient of inflation rate (-274.07) is negative and significant at lag 2 which implies that past  
inflation rate has a negative and significant relationship with GDP. This implies that decrease in inflation two  
periods ago is associated with an increase in GDP. That is, lower inflation levels in the recent past have a  
positive effect on economic output.  
Page 3660  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
Table 4: Long run ARDL  
Conditional Error Correction Regression  
Variable  
Coefficient  
Std. Error  
t-Statistic  
Prob.  
C
GDP(-1)*  
NPI(-1)  
NDI(-1)  
EXRATE(-1)  
INF(-1)  
D(GDP(-1))  
D(GDP(-2))  
D(GDP(-3))  
D(NPI)  
D(NPI(-1))  
D(NPI(-2))  
D(NPI(-3))  
D(NDI)  
D(NDI(-1))  
D(NDI(-2))  
D(NDI(-3))  
D(EXRATE)  
D(EXRATE(-1))  
D(EXRATE(-2))  
D(EXRATE(-3))  
D(INF)  
-3260.083  
-0.247472  
-0.144496  
-0.130334  
722.5001  
187.7960  
-0.013927  
-0.154846  
-0.841945  
-0.091612  
0.071996  
0.033161  
0.021692  
-0.016124  
0.048507  
0.039730  
0.133823  
55.55277  
-333.3141  
-330.7031  
-155.0810  
-199.0215  
-106.1070  
-274.0704  
-76.76570  
7382.140  
0.056334  
0.025544  
0.035110  
104.9756  
169.7986  
0.179890  
0.187353  
0.203732  
0.016445  
0.015816  
0.017758  
0.016908  
0.013569  
0.033001  
0.027972  
0.026929  
83.35716  
105.1907  
92.74182  
81.89616  
102.1823  
152.2080  
102.4690  
92.98453  
-0.441618  
-4.392927  
-5.656854  
-3.712118  
6.882553  
1.105993  
-0.077420  
-0.826496  
-4.132613  
-5.570888  
4.551987  
1.867358  
1.282953  
-1.188256  
1.469858  
1.420349  
4.969475  
0.666443  
-3.168664  
-3.565847  
-1.893629  
-1.947710  
-0.697118  
-2.674667  
-0.825575  
0.6705  
0.0023  
0.0005  
0.0059  
0.0001  
0.3009  
0.9402  
0.4325  
0.0033  
0.0005  
0.0019  
0.0988  
0.2354  
0.2688  
0.1798  
0.1933  
0.0011  
0.5239  
0.0132  
0.0073  
0.0949  
0.0873  
0.5055  
0.0282  
0.4330  
D(INF(-1))  
D(INF(-2))  
D(INF(-3))  
Case 2: Restricted Constant and No Trend  
Variable  
Coefficient  
Std. Error  
t-Statistic  
Prob.  
NPI  
NDI  
EXRATE  
INF  
-0.583889  
-0.526661  
2919.527  
758.8588  
-13173.56  
0.113563  
0.078476  
373.7459  
621.2102  
28489.17  
-5.141523  
-6.711075  
7.811528  
1.221581  
-0.462406  
0.0009  
0.0002  
0.0001  
0.2566  
0.6561  
C
EC = GDP - (-0.5839*NPI -0.5267*NDI + 2919.5267*EXRATE + 758.8588*INF  
- 13173.5618)  
The result presented in Table 4 (Conditional Error Correction Long-run Regression result for economic  
development) indicates that in the long run, a year lag of (GDP) is significant at 5% level.  
The coefficient of NPI (Net Private Investment) to GDP is -0.583 and statistically significant in the long run, it  
suggests a robust negative relationship between net private investment and economic development over  
extended periods. This negative coefficient implies that as net private investment increases, GDP or economic  
development decreases in the long run. This could be as a result of underlying structural weaknesses in the  
economy that prevent private investment from translating into sustainable economic growth. These weaknesses  
may include inadequate infrastructure, regulatory barriers, or institutional deficiencies. This can be resolved by  
implementing structural reforms to address inefficiencies and improve the business environment, thereby  
enhancing the effectiveness of private investment in driving economic development. Linking to the integrated  
financial liberalization theory, this negative effect aligns with the argument that unchecked liberalization in  
economies with low levels of financial development can trigger volatility and asset bubbles (Stiglitz, 2000;  
Stiglitz & Weiss, 1981).  
The coefficient of NDI (Net Direct Investment) to GDP is -0.526 and statistically significant in the long run. It  
implies that changes in net direct investment levels are inversely associated with changes in GDP. That is, an  
increase in NDI is associated with a decrease in GDP. Dependence on NDI as a driver of economic growth  
may pose risks, especially if it exhibits a negative relationship with GDP. Policymakers should consider  
diversifying sources of investment and other forms of foreign investment to reduce vulnerability to fluctuations  
in NDI. This finding resonates with internalization theory, which argues that market imperfections, such as  
imperfect information and high transaction costs limit the ability of firms to internalize the benefits of FDI, a  
challenge evident in Nigeria’s institutional environment (Hymer, 1976; Dunning, 1988)  
The coefficient of the exchange rate to GDP is 2919.527 and statistically significant in the long run, it implies  
a strong positive relationship between the exchange rate and economic development. This positive coefficient  
suggests that an increase in the exchange rate is associated with an increase in GDP or economic development  
in the long run. This implies that, a higher exchange rate may attract foreign investment by making domestic  
assets more attractive to foreign investors. This influx of investment can stimulate economic activity and  
contribute to development but this is not the situation in Nigeria as depreciation in Naira, as make import  
Page 3661  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
goods more expensive as the country is characterized by over dependences on importation. Policymakers may  
aim to maintain a stable exchange rate to provide certainty for businesses and investors, thereby supporting  
long-term economic development.  
The coefficient of inflation to GDP is 758.85 but statistically insignificant in the long run, it suggests a lack of  
robust relationship between inflation and economic development over extended periods. The insignificant  
coefficient implies that changes in inflation do not have a statistically significant impact on GDP or economic  
development in the long run. While inflation may not directly impact economic development in the long run,  
maintaining price stability remains important for preserving consumer purchasing power, fostering confidence,  
and supporting overall economic stability.  
The negative effects of capital flows (NPI and NDI) on development in Nigeria can be attributed to several  
factors. First, capital flight depletes domestic investment resources, leaving fewer funds for productive  
activities and leading to reduced GDP growth (Muhammad et al., 2023). Second, high volatility in inflows,  
driven by external factors like oil prices and global risk aversion, causes currency depreciation and reserve  
depletion, exacerbating economic instability (IMF, 2016). Third, foreign debt components of capital flows  
create debt overhang, with servicing costs diverting resources from investment (Okolie & Ruth, 2021).  
Additionally, poor institutional quality and corruption facilitate illicit outflows, further hindering growth  
(Muhammad et al., 2023). Nigeria's oil dependency amplifies "Dutch disease," where inflows appreciate the  
currency, undermining non-oil sectors (Prasad et al., 2006). Weak absorptive capacity due to underdeveloped  
financial systems prevents efficient allocation of inflows, aligning with the integrated theory's warnings on  
liberalization risks in low-development contexts.  
Diagnostic Tests  
The Diagnostic tests for serial correlation, Heteroskedasticity and normality were conducted, and the results  
are presented in Table 5. The result for serial correlation shows that errors in equation are not serially  
correlated. The test for Heteroskedasticity indicated that there were equal spreads in variance in the equations  
of the model with a probability value of 0.8967. The normality test of the equation in the model shows that the  
equation allows the normal distribution. The Jarque – Bera statistics of 0.699525 with the probability values of  
0.704856 shows that the variables in the model are normally distributed.  
Table 5: Diagnostic Tests for Model 1  
Test  
R2 Statistics  
Probability Value  
0.3748  
Breusch-Godfrey LM test for 1.962815  
Serial correlation  
White Heteroskedasticity  
Normality Test  
15.75495  
0.699525  
0.8967  
0.704856  
Stability Test  
It is ideal to investigate the stability of ARDL model. For this purpose, we have checked the stability of the  
model parameters using both cumulative sum of recursive residuals (CUSUM) and the cumulative sum of  
squares of recursive residuals (CUSUMSQ) test procedures. CUSUM and (CUSUMSQ) are plotted against the  
break points. The plot of the CUSUM and (CUSUMSQ) are obtained from a recursive estimation of the  
model.The graph (figure 3 and 4) below depicts the results for CUSUM and (CUSUMSQ) test. The results  
indicate stability in the coefficients of the model, because the plots of the CUSUM statistic fall inside the  
critical bounds of 5% confidence interval of parameter stability.  
Page 3662  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
Figure 3: CUSUM  
10.0  
7.5  
5.0  
2.5  
0.0  
-2.5  
-5.0  
-7.5  
-10.0  
2015  
2016  
2017  
2018  
2019  
2020  
2021  
2022  
CUSUM  
5% Significance  
Figure 4: CUSUM of Squares  
1.6  
1.2  
0.8  
0.4  
0.0  
-0.4  
2015  
2016  
2017  
2018  
2019  
2020  
2021  
2022  
CUSUM of Squares  
5% Significance  
Model 2 on financial stability and economic development  
Table 6 ARDL Bond Test Result for Model 2  
Model  
F Statistics  
8.45  
5% Critical value  
Decision  
Equation 1  
Co- integration  
I(0)  
I(1)  
2.56  
3.49  
Source: Authors’ computation using (EVIEWS 12)  
The ARDL bounds test is based on the assumption that the variables are I (0) or I (1) as shown above in the  
unit root table. The results of the ARDL bounds testing approach are presented in table (2) indicating that the  
computed F-statistics for explanatory variables was (8.45). The f- bound test statistics of (8.45) exceeds upper  
critical bound (3.49) at 5% level of significance. This finding supports that co-integrating relationship exists  
and confirms the stable long-run relationship between the variables. This implies that the null hypothesis of no  
co-integration among the variables is rejected.  
Page 3663  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
Table 7: ARDL Error Correction (ECM) Regression for Model 2  
ECM Regression  
Case 2: Restricted Constant and No Trend  
Variable  
Coefficient  
Std. Error  
t-Statistic  
Prob.  
D(CPS)  
D(CPS(-1))  
D(CPS(-2))  
D(NIM)  
D(NIM(-1))  
D(NIM(-2))  
D(NIM(-3))  
D(EXRATE)  
D(EXRATE(-1))  
D(EXRATE(-2))  
D(INF)  
1.981879  
1.693789  
8.827085  
1750.944  
1620.342  
1868.678  
3251.939  
-144.4619  
-226.8476  
-143.2131  
-128.4055  
-0.081407  
1.634461  
1.643231  
1.732280  
1058.148  
1048.998  
955.1470  
951.8353  
64.46935  
76.31885  
82.75474  
117.2323  
0.009982  
1.212559  
1.030768  
5.095646  
1.654725  
1.544657  
1.956430  
3.416493  
-2.240785  
-2.972366  
-1.730573  
-1.095308  
-8.155407  
0.2429  
0.3180  
0.0001  
0.1175  
0.1420  
0.0681  
0.0035  
0.0396  
0.0090  
0.1028  
0.2896  
0.0000  
CointEq(-1)*  
R-squared  
0.999373 Mean dependent var  
0.998746 S.D. dependent var  
9814.671 Akaike info criterion  
1.54E+09 Schwarz criterion  
-338.2042 Hannan-Quinn criter.  
1594.389 Durbin-Watson stat  
0.000000  
286690.9  
277200.5  
21.52753  
22.29846  
21.78692  
2.206306  
Adjusted R-squared  
S.E. of regression  
Sum squared resid  
Log likelihood  
F-statistic  
Prob(F-statistic)  
The result showed that R2 (R-Squared) value of 0.999 was obtained. This implies that 99% of the variations in  
economic development is explained by changes financial stability variables. The F-Statistics of 1594.389 is  
significant considering the probability value. This implies that the model has goodness of fit. The Durbin-  
Watson value of 2.20 which is approximately 2 indicates that there exists no serial correlation. The error  
correction model (ECM) for economic development and financial stability is specified, following the  
Representation Theorem specified. The primary reason for this is to capture the dynamics in economic  
development and financial stability in the short-run and identify the speed of adjustment as a response to  
departures from the long-run equilibrium. Thus, the general specification framework of the ECM is shown in  
Table 7.  
Statistically, the ECM term is negative and significant at 5% probability level. The existence of short run  
equilibrium among the time series in government expenditure on agriculture equation is validated by this  
result. The slope coefficient of the error term in absolute terms (8.14) represents the speed of adjustment and is  
consistent with the hypothesis of convergence towards the long-run equilibrium once economic development  
equation fluctuates from its equilibrium in the short run. The coefficient of the ECMt-1 suggests that economic  
development adjusts to the explanatory time series as about 8.1 percent of the discrepancy between long-run  
and short-run is corrected annually in the country.  
In the short run, the coefficient of credit to private sector (1.98), (1.69), (8.82) at current level, lag 1 and lag 2  
respectively. The result indicates that CPS (credit to the private sector) is positive but statistically insignificant  
in the current period and lagged one period to GDP, while lagged two periods is positive and statistically  
significant to GDP, it suggests a delayed impact of changes in credit to the private sector on GDP. The positive  
coefficient for CPS in the current period indicates that an increase in credit to the private sector is associated  
with an increase in GDP, but this relationship is not statistically significant. However, the positive and  
significant coefficient for CPS lagged two periods suggests that changes in credit to the private sector two  
periods ago have a significant positive impact on current GDP. Policymakers may need to consider the timing  
and magnitude of monetary policy interventions, such as interest rate adjustments or liquidity injections, in  
response to changes in credit to the private sector. While current changes in credit may not have an immediate  
impact on GDP, the significant lagged effect suggests that policy responses may need to be implemented with  
a delay.  
The NIM (Net Interest Margin) is positive but statistically insignificant in the current period and lagged one  
period to GDP, while lagged three periods is positive and statistically significant to GDP, it suggests a delayed  
impact of changes in net interest margin on GDP. The positive coefficient for NIM in the current period  
Page 3664  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
indicates that an increase in net interest margin is associated with an increase in GDP, but this relationship is  
not statistically significant. However, the positive and significant coefficient for NIM lagged three periods  
suggests that changes in net interest margin three periods ago have a significant positive impact on current  
GDP. The policy response would involve a nuanced approach that considers both the immediate and delayed  
effects of changes in net interest margin on GDP, ensuring that policy interventions are timely and effective in  
supporting overall economic stability and growth.  
The coefficient of exchange rate is negative and statistically significant in the current period and lagged one  
period to GDP, it suggests an immediate impact of changes in the exchange rate on GDP. The negative  
coefficient for the exchange rate in the current period indicates that a depreciation of the domestic currency  
(increase in the exchange rate) is associated with a decrease in GDP, and this relationship is statistically  
significant. This implies that a sudden depreciation of the domestic currency may negatively affect economic  
output in the short run.  
The negative coefficient for inflation in the current period indicates that a decrease in inflation (or deflation) is  
associated with a decrease in GDP, but this relationship is not statistically significant. This implies that  
changes in inflation levels are not having a statistically significant effect on economic output in the short run.  
Table 8: Long run ARDL  
Conditional Error Correction Regression  
Variable  
Coefficient  
Std. Error  
t-Statistic  
Prob.  
C
-23460.67  
-0.081407  
-0.070579  
2369.317  
344.5306  
43.79755  
1.981879  
1.693789  
8.827085  
1750.944  
1620.342  
1868.678  
3251.939  
-144.4619  
-226.8476  
-143.2131  
-128.4055  
15303.09  
0.081695  
1.846356  
2007.776  
70.28809  
148.6213  
2.195619  
2.215558  
2.203769  
1450.578  
1540.099  
1214.529  
1278.392  
88.11796  
114.8336  
109.5560  
152.0139  
-1.533068  
-0.996483  
-0.038226  
1.180070  
4.901693  
0.294692  
0.902652  
0.764498  
4.005450  
1.207067  
1.052102  
1.538603  
2.543772  
-1.639415  
-1.975446  
-1.307213  
-0.844696  
0.1448  
0.3338  
0.9700  
0.2552  
0.0002  
0.7720  
0.3801  
0.4557  
0.0010  
0.2449  
0.3084  
0.1434  
0.0217  
0.1206  
0.0657  
0.2096  
0.4107  
GDP(-1)*  
CPS(-1)  
NIM(-1)  
EXRATE(-1)  
INF(-1)  
D(CPS)  
D(CPS(-1))  
D(CPS(-2))  
D(NIM)  
D(NIM(-1))  
D(NIM(-2))  
D(NIM(-3))  
D(EXRATE)  
D(EXRATE(-1))  
D(EXRATE(-2))  
D(INF)  
Case 2: Restricted Constant and No Trend  
Variable  
Coefficient  
Std. Error  
t-Statistic  
Prob.  
CPS  
NIM  
EXRATE  
INF  
-0.866980  
29104.47  
4232.181  
538.0049  
-288188.7  
23.48368  
28443.15  
3927.166  
2127.282  
325450.4  
-0.036918  
1.023251  
1.077668  
0.252907  
-0.885507  
0.9710  
0.3214  
0.2972  
0.8036  
0.3890  
C
EC = GDP - (-0.8670*CPS + 29104.4684*NIM + 4232.1810*EXRATE +  
538.0049*INF - 288188.6825)  
The result presented in Table 8 (Conditional Error Correction Long-run Regression result for economic  
development) indicates that in the long run, a year lag of (EXRATE) is positive and significant at 5% level to  
GDP. It implies that, a postive exchange rate could attract foreign investment, which could contributes to  
economic development.  
The coefficient of Credit to private sector (CPS) at lag 2 is positive and significant to GDP in the long run. It  
implies that, an increase in CPS is associated with a positive effect on GDP in the current period. That is,  
increased credit to the private can stimulate investment and consumption, leading to higher economic activity  
and development  
Net interest margin (NIM) indicates a positive and significant relationship with GDP, which implies that, as  
the economy grows, there is an associated increase in NIM. This indicates that financial institutions are  
effectively managing their interest rate spread, which can lead to higher profits  
While, (case 2 restricted constant and no trend) shows that in the long-run CPS is negative to GDP, Net  
interest margin (NIM), EXRATE, INF is all positive. The variables are not insignificant in the long-run. This  
implies that, an increase in credit extended to the private sector does not significantly contribute to long-term  
Page 3665  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
economic development. Positive but insignificant relationship for NIM, Exchange rate and inflation, suggests  
that changes in these factors do not have a significant long-term impact on GDP. These findings align with the  
integrated financial liberalization theory, which argues that weak financial development, such as that observed  
in Nigeria results in inefficient credit allocation and consequently negative or insignificant economic outcomes  
(Rodrik, 1998; Chang, 2002)  
Diagnostic Tests  
The Diagnostic tests for serial correlation, Heteroskedasticity and normality were conducted, and the results  
are presented in Table 5. The result for serial correlation shows that errors in equation are not serially  
correlated. The test for Heteroskedasticity indicated that there were equal spreads in variance in the equations  
of the model with a probability value of 0.6016. The normality test of the equation in the model shows that the  
equation allows the normal distribution. The Jarque – Bera statistics of 1.089032 with the probability values of  
0.580123 shows that the variables in the model are normally distributed.  
Table 9: Diagnostic Tests for Model II  
Test  
R2 Statistics  
Probability Value  
0.6172  
Breusch-Godfrey LM test for 0.965232  
Serial correlation  
White Heteroskedasticity  
Normality Test  
13.96071  
1.089032  
0.6016  
0.580123  
Stability Test  
It is ideal to investigate the stability of ARDL model. For this purpose, we have checked the stability of the  
model parameters using both cumulative sum of recursive residuals (CUSUM) and the cumulative sum of  
squares of recursive residuals (CUSUMSQ) test procedures. CUSUM and (CUSUMSQ) are plotted against the  
break points. The plot of the CUSUM and (CUSUMSQ) are obtained from a recursive estimation of the model.  
The graph (figure 5 and 6) below depicts the results for CUSUM and (CUSUMSQ) test. The results indicate  
stability in the coefficients of the model, because the plots of the CUSUM statistic fall inside the critical  
bounds of 5% confidence interval of parameter stability.  
Figure 5  
12  
8
4
0
-4  
-8  
-12  
2008  
2010  
2012  
2014  
2016  
2018  
2020  
2022  
CUSUM  
5% Significance  
Page 3666  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
Figure 6  
1.6  
1.2  
0.8  
0.4  
0.0  
-0.4  
2008  
2010  
2012  
2014  
2016  
2018  
2020  
2022  
CUSUM of Squares  
5% Significance  
SUMMARY, CONCLUSION AND RECOMMENDATIONS  
The purpose of this study was to examine impact of capital outflow and financial sector stability on economic  
development in Nigeria by using data extending over a period of 1986 to 2022. The unit root test was  
conducted to ascertain the stationarity of the variables’ the ARDL Error Correction regression analysis was  
used to test the short run relationship capital outflow and financial sector stability on economic development.  
The ARDL long run analysis was used to ascertain the long-run relationship among the variables. This study  
has shown that the variables for the model were normally distributed.  
The result from the capital outflow variables showed that both net port-folio investment and net direct  
investment indicated a negative but significant relationship with GDP, this negative relationship highlights  
potential challenges in leveraging foreign capital to drive sustained economic development. It suggests that  
simply increasing foreign investment inflows may not guarantee positive economic outcomes.  
For financial sector stability it was discovered that, in the long-run CPS is negative to GDP, Net interest  
margin (NIM), EXRATE, INF are all positive but all the variables are not insignificant in the long-run. This  
calls for the need to reassess the effectiveness of credit allocation policies to ensure that credit expansion to the  
private sector aligns with long-term economic development objectives. Also, given the insignificant effect of  
the variables others factors such as investment, innovation, and productivity should be considered to stimulate  
economic development. The result conform from the integrated financial liberalization and financial repression  
theory financial markets may become more prone to speculative excesses, leading to boom-bust cycles and  
financial crises and undermining the objective of economic development in the economy.  
To situate Nigeria’s experience within broader global patterns, comparative evidence from emerging  
economies reveals mixed outcomes. In South Africa, capital flows have been volatile due to heavy commodity  
dependence similar to Nigeria’s reliance on oil which has contributed to periodic outflows and growth  
slowdowns (Brooks, 2025; ISS, 2025). Although both countries face commodity-induced volatility, South  
Africa’s relatively stronger institutional framework mitigates some risks, whereas Nigeria’s weaker  
governance structures exacerbate capital flight (Muhammad et al., 2023). In Brazil, episodes of large capital  
inflows have led to currency appreciation and symptoms of “Dutch disease,” negatively affecting  
manufacturing productivity, a challenge that mirrors Nigeria’s non-oil sector struggles (Prasad et al., 2006).  
India presents a contrasting case, where robust financial development and higher absorptive capacity have  
enabled FDI to stimulate manufacturing and overall growth (End, 2024; ScienceDirect, 2024). In Kenya, while  
remittances and FDI support growth, evidence shows that aid generates negative long-run effects, reflecting  
broader patterns observed across Sub-Saharan Africa, including Nigeria (Taylor and Francis, 2025; ISS, 2025).  
Overall, Nigeria’s negative capital-flow effects are consistent with outcomes in other commodity-dependent  
economies but are intensified by institutional weaknesses, underscoring the urgency of reforms that emulate  
the successes observed in more financially developed economies such as India.  
Page 3667  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
RECOMMENDATION  
From the findings of this study, the following recommendations are given;  
1. Give the significant effect of capital flow, policymakers should aim to strike a balance between attracting  
foreign investment and ensuring that it aligns with broader economic development goals. This may  
involve fostering a conducive investment environment, enhancing regulatory frameworks, and promoting  
domestic capacity-building initiatives.  
2. There should be implementation of targeted policies to improve credit allocation efficiency, ensuring that  
credit is directed towards productive sectors that contribute to economic development.  
REFERENCES  
1. Abbes, S. M., Mostefa, B., Seghir, G. M., and Zakarya, G. Y. (2015). Causal interactions between FDI  
and economic growth: Evidence from dynamic panel co-integration. Procedia Economics and Finance,  
23, 276–290. https://doi.org/10.1016/S2212-5671(15)00358-8  
2. Aizenman, J., Jinjarak, Y., and Park, D. (2011). Capital flows and economic growth in the era of  
financial integration and crisis, 1990–2010. NBER Working Paper No. 17502. National Bureau of  
Economic Research.  
3. Albulescu, C. T. (2015). Do foreign direct and portfolio investments affect long-term economic growth  
in Central and Eastern Europe? Procedia Economics and Finance, 23, 507–512.  
https://doi.org/10.1016/S2212-5671(15)00580-0  
4. Alfaro, L., Kalemli-Ozcan, S., and Volosovych, V. (2007). Capital flows in a globalized world: The role  
of policies and institutions. In S. Edwards (Ed.), Capital controls and capital flows in  
economies: Policies, practices and consequences (pp. 19–72). University of Chicago  
emerging  
Press.  
5. Bekaert, G., and Harvey, C. R. (2000). Capital flows and the behavior of emerging market equity return.  
In S. Edwards (Ed.), Capital flows and the emerging economies: Theory, evidence, and controversies  
(pp. 159–194). University of Chicago Press.  
6. Brooks, C. (2025). Capital flow volatility and macroeconomic performance in commodity-dependent  
economies: The case of South Africa. Journal of African Economies (forthcoming).  
7. Brown, R. L., Durbin, J., and Evans, J. M. (1975). Techniques for testing the constancy of regression  
relationships over time. Journal of the Royal Statistical Society: Series B (Methodological), 37(2), 149–  
163.  
8. Central Bank of Nigeria (CBN). (1997). Central Bank of Nigeria statistical bulletin. Abuja: CBN.  
9. Central Bank of Nigeria (CBN). (2022). Central Bank of Nigeria statistical bulletin. Abuja: CBN.  
10. Choong, C.-K., Yusop, Z., and Law, S.-H. (2010). Private capital flows to developing countries: The  
role of the domestic financial sector. Journal of the Asia Pacific Economy, 15(4), 508–529.  
https://doi.org/10.1080/13547860.2010.516170  
11. De Vita, G., and Kyaw, K. S. (2008). Determinants of capital flows to developing countries: A structural  
VAR analysis. Journal of Economic Studies, 35(4), 304–322.  
12. Edwards, S. (2001). Capital mobility and economic performance: Are emerging economies different? In  
H. Siebert (Ed.), The world’s new financial landscape: Challenges for economic policy (pp. 219–244).  
Springer-Verlag.  
13. Eichengreen, B. (2000). Taming capital flows. World Development, 28(6), 1105–1116.  
14. End, N. (2024). Capital flows and growth in emerging Asia: Does financial development matter? Asian  
Development Review, 41(1), 45–78.  
15. Frankel, J. A. (1992). Measuring international capital mobility: A review. American Economic Review,  
82(2), 197–202.  
16. Harley, T. W., and Akinola, A. T. (2018). The impact of capital flows on financial system stability in  
Nigeria. African Journal of Accounting and Financial Research, 1(1), 1–8.  
17. Harris, R., and Sollis, R. (2003). Applied time series modelling and forecasting. Wiley.  
18. International Monetary Fund (IMF). (2016). Capital flows: Review of experience with the institutional  
view. IMF Policy Paper.  
Page 3668  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
19. Justine, C. (2018). The effect of capital flows on key financial stability measures in Jamaica. Financial  
Stability Department, Bank of Jamaica.  
20. Kose, M. A., Prasad, E., Rogoff, K., and Wei, S.-J. (2010). Financial globalization and economic  
policies. In D. Rodrik & M. Rosenzweig (Eds.), Handbook of development economics (Vol. 5, pp.4283–  
4359). North-Holland.  
21. Mohan, R., and Kapur, M. (2010). Liberalization and regulation of capital flows: Lessons for emerging  
market economies. ADBI Working Paper Series No. 186. Asian Development Bank Institute.  
22. Muhammad, A., Usman, M., and Adebayo, T. S. (2023). Capital flight and economic growth in Sub-  
Saharan Africa: The role of institutional quality. Resources Policy, 80, 103–120.  
23. Nyang`oro, O. (2017). Capital inflows and economic growth in Sub-Saharan African countries. African  
Development Bank Working Paper Series No. 285.  
24. Obadan, M. I. (2004). Globalization and economic management in Africa’s economic development. In  
Nigeria Economic Society Annual Conference Proceedings (pp. 3–34). Ibadan.  
25. Okolie, P. I., and Ruth, O. (2021). External debt and economic growth in Nigeria: A threshold analysis.  
CBN Journal of Applied Statistics, 12(1), 45–68.  
26. Onosode, G. O. (1997). The Nigerian capital market in the 1990s. CBN Economic and Financial  
Review, 35(4).  
27. Pesaran, M. H., Shin, Y., and Smith, R. J. (2001). Bounds testing approaches to the analysis of level  
relationships. Journal of Applied Econometrics, 16(3), 289–326.  
28. Prasad, E. S., Rajan, R. G., and Subramanian, A. (2006). Foreign capital and economic growth.  
Brookings Papers on Economic Activity, 2007(1), 153–230.  
Page 3669  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
Appendix  
Gdp At First Difference  
Exogenous: Constant, Linear Trend  
Lag Length: 0 (Automatic - based on SIC, maxlag=9)  
t-Statistic  
Prob.*  
Augmented Dickey-Fuller test statistic  
-4.158658  
-4.243644  
-3.544284  
-3.204699  
0.0123  
Test critical values:  
1% level  
5% level  
10% level  
*MacKinnon (1996) one-sided p-values.  
Augmented Dickey-Fuller Test Equation  
Dependent Variable: D(GDP,2)  
Method: Least Squares  
Date: 02/19/24 Time: 21:36  
Sample (adjusted): 1988 2022  
Included observations: 35 after adjustments  
Variable  
Coefficient  
Std. Error  
t-Statistic  
Prob.  
D(GDP(-1))  
C
@TREND("1986")  
-0.835498  
-10499.45  
1737.344  
0.200906  
5729.447  
435.0075  
-4.158658  
-1.832542  
3.993825  
0.0002  
0.0762  
0.0004  
R-squared  
0.364146 Mean dependent var  
0.324406 S.D. dependent var  
14843.64 Akaike info criterion  
7.05E+09 Schwarz criterion  
-384.2811 Hannan-Quinn criter.  
9.163027 Durbin-Watson stat  
0.000714  
2869.145  
18059.16  
22.13035  
22.26366  
22.17637  
1.820292  
Adjusted R-squared  
S.E. of regression  
Sum squared resid  
Log likelihood  
F-statistic  
Prob(F-statistic)  
Cps At First Difference  
Exogenous: Constant, Linear Trend  
Lag Length: 0 (Automatic - based on SIC, maxlag=9)  
t-Statistic  
Prob.*  
Augmented Dickey-Fuller test statistic  
-3.912883  
-4.243644  
-3.544284  
-3.204699  
0.0220  
Test critical values:  
1% level  
5% level  
10% level  
*MacKinnon (1996) one-sided p-values.  
Augmented Dickey-Fuller Test Equation  
Dependent Variable: D(CPS,2)  
Method: Least Squares  
Date: 02/19/24 Time: 21:39  
Sample (adjusted): 1988 2022  
Included observations: 35 after adjustments  
Variable  
Coefficient  
Std. Error  
t-Statistic  
Prob.  
D(CPS(-1))  
C
@TREND("1986")  
-0.782216  
-811.4460  
90.49926  
0.199908  
437.6133  
26.21056  
-3.912883  
-1.854253  
3.452778  
0.0004  
0.0729  
0.0016  
R-squared  
0.338943 Mean dependent var  
0.297627 S.D. dependent var  
1136.140 Akaike info criterion  
41306030 Schwarz criterion  
-294.3333 Hannan-Quinn criter.  
8.203668 Durbin-Watson stat  
0.001330  
174.3123  
1355.651  
16.99048  
17.12379  
17.03650  
1.814420  
Adjusted R-squared  
S.E. of regression  
Sum squared resid  
Log likelihood  
F-statistic  
Prob(F-statistic)  
Page 3670  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
Nim At Level  
Null Hypothesis: NIM has a unit root  
Exogenous: Constant, Linear Trend  
Lag Length: 0 (Automatic - based on SIC, maxlag=9)  
t-Statistic  
Prob.*  
Augmented Dickey-Fuller test statistic  
-3.550393  
-4.234972  
-3.540328  
-3.202445  
0.0489  
Test critical values:  
1% level  
5% level  
10% level  
*MacKinnon (1996) one-sided p-values.  
Augmented Dickey-Fuller Test Equation  
Dependent Variable: D(NIM)  
Method: Least Squares  
Date: 02/19/24 Time: 21:41  
Sample (adjusted): 1987 2022  
Included observations: 36 after adjustments  
Variable  
Coefficient  
Std. Error  
t-Statistic  
Prob.  
NIM(-1)  
C
@TREND("1986")  
-0.481761  
3.210213  
0.017962  
0.135692  
0.878702  
0.029478  
-3.550393  
3.653358  
0.609346  
0.0012  
0.0009  
0.5465  
R-squared  
0.301512 Mean dependent var  
0.259180 S.D. dependent var  
1.605062 Akaike info criterion  
85.01539 Schwarz criterion  
-66.54943 Hannan-Quinn criter.  
7.122458 Durbin-Watson stat  
0.002683  
0.156862  
1.864813  
3.863857  
3.995817  
3.909915  
1.928744  
Adjusted R-squared  
S.E. of regression  
Sum squared resid  
Log likelihood  
F-statistic  
Prob(F-statistic)  
Exchange Rate At First Difference  
Exogenous: Constant, Linear Trend  
Lag Length: 0 (Automatic - based on SIC, maxlag=9)  
t-Statistic  
Prob.*  
0.0007  
Augmented Dickey-Fuller test statistic  
-5.281928  
-4.243644  
-3.544284  
-3.204699  
Test critical values:  
1% level  
5% level  
10% level  
*MacKinnon (1996) one-sided p-values.  
Augmented Dickey-Fuller Test Equation  
Dependent Variable: D(EXRATE,2)  
Method: Least Squares  
Date: 02/19/24 Time: 21:43  
Sample (adjusted): 1988 2022  
Included observations: 35 after adjustments  
Variable  
Coefficient  
Std. Error  
t-Statistic  
Prob.  
D(EXRATE(-1))  
C
@TREND("1986")  
-0.931423  
-4.188371  
0.861350  
0.176341  
8.954916  
0.444780  
-5.281928  
-0.467717  
1.936577  
0.0000  
0.6432  
0.0617  
R-squared  
0.465766 Mean dependent var  
0.432376 S.D. dependent var  
24.73600 Akaike info criterion  
19579.83 Schwarz criterion  
-160.3837 Hannan-Quinn criter.  
13.94940 Durbin-Watson stat  
0.000044  
0.689286  
32.83213  
9.336213  
9.469528  
9.382233  
1.972680  
Adjusted R-squared  
S.E. of regression  
Sum squared resid  
Log likelihood  
F-statistic  
Prob(F-statistic)  
Page 3671  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
Inflation At First Difference  
Exogenous: Constant, Linear Trend  
Lag Length: 0 (Automatic - based on SIC, maxlag=9)  
t-Statistic  
Prob.*  
Augmented Dickey-Fuller test statistic  
-5.604783  
-4.243644  
-3.544284  
-3.204699  
0.0003  
Test critical values:  
1% level  
5% level  
10% level  
*MacKinnon (1996) one-sided p-values.  
Augmented Dickey-Fuller Test Equation  
Dependent Variable: D(INF,2)  
Method: Least Squares  
Date: 02/19/24 Time: 21:47  
Sample (adjusted): 1988 2022  
Included observations: 35 after adjustments  
Variable  
Coefficient  
Std. Error  
t-Statistic  
Prob.  
D(INF(-1))  
C
@TREND("1986")  
-0.991603  
1.171382  
-0.044159  
0.176921  
6.057646  
0.281541  
-5.604783  
0.193372  
-0.156847  
0.0000  
0.8479  
0.8764  
R-squared  
0.495381 Mean dependent var  
0.463842 S.D. dependent var  
16.81324 Akaike info criterion  
9045.920 Schwarz criterion  
-146.8705 Hannan-Quinn criter.  
15.70708 Durbin-Watson stat  
0.000018  
0.277057  
22.96175  
8.564027  
8.697342  
8.610047  
1.650674  
Adjusted R-squared  
S.E. of regression  
Sum squared resid  
Log likelihood  
F-statistic  
Prob(F-statistic)  
Npi At First Difference  
Exogenous: Constant, Linear Trend  
Lag Length: 0 (Automatic - based on SIC, maxlag=9)  
t-Statistic  
Prob.*  
Augmented Dickey-Fuller test statistic  
-6.439144  
-4.243644  
-3.544284  
-3.204699  
0.0000  
Test critical values:  
1% level  
5% level  
10% level  
*MacKinnon (1996) one-sided p-values.  
Augmented Dickey-Fuller Test Equation  
Dependent Variable: D(NPI,2)  
Method: Least Squares  
Date: 02/19/24 Time: 22:19  
Sample (adjusted): 1988 2022  
Included observations: 35 after adjustments  
Variable  
Coefficient  
Std. Error  
t-Statistic  
Prob.  
D(NPI(-1))  
C
@TREND("1986")  
-1.129526  
-2029.612  
-402.6020  
0.175416  
50743.86  
2359.980  
-6.439144  
-0.039997  
-0.170596  
0.0000  
0.9683  
0.8656  
R-squared  
0.564438 Mean dependent var  
0.537215 S.D. dependent var  
140904.5 Akaike info criterion  
6.35E+11 Schwarz criterion  
-463.0490 Hannan-Quinn criter.  
20.73414 Durbin-Watson stat  
0.000002  
303.0543  
207126.3  
26.63137  
26.76468  
26.67739  
2.053330  
Adjusted R-squared  
S.E. of regression  
Sum squared resid  
Log likelihood  
F-statistic  
Prob(F-statistic)  
Page 3672  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
Ndi At First Difference  
Exogenous: Constant, Linear Trend  
Lag Length: 3 (Automatic - based on SIC, maxlag=9)  
t-Statistic  
Prob.*  
Augmented Dickey-Fuller test statistic  
-4.296517  
-4.273277  
-3.557759  
-3.212361  
0.0095  
Test critical values:  
1% level  
5% level  
10% level  
*MacKinnon (1996) one-sided p-values.  
Augmented Dickey-Fuller Test Equation  
Dependent Variable: D(NDI,2)  
Method: Least Squares  
Date: 02/19/24 Time: 22:21  
Sample (adjusted): 1991 2022  
Included observations: 32 after adjustments  
Variable  
Coefficient  
Std. Error  
t-Statistic  
Prob.  
D(NDI(-1))  
D(NDI(-1),2)  
D(NDI(-2),2)  
D(NDI(-3),2)  
C
-2.587230  
1.413903  
0.509508  
0.874659  
36930.86  
-2876.496  
0.602169  
0.575128  
0.493147  
0.346340  
62149.82  
2894.895  
-4.296517  
2.458415  
1.033175  
2.525434  
0.594223  
-0.993644  
0.0002  
0.0209  
0.3110  
0.0180  
0.5575  
0.3296  
@TREND("1986")  
R-squared  
0.690894 Mean dependent var  
0.631450 S.D. dependent var  
139671.2 Akaike info criterion  
5.07E+11 Schwarz criterion  
-421.1893 Hannan-Quinn criter.  
11.62270 Durbin-Watson stat  
0.000006  
24415.01  
230069.5  
26.69933  
26.97416  
26.79043  
2.015559  
Adjusted R-squared  
S.E. of regression  
Sum squared resid  
Log likelihood  
F-statistic  
Prob(F-statistic)  
Model One  
Ardl Estimation  
Method: ARDL  
Date: 02/19/24 Time: 21:49  
Sample (adjusted): 1990 2022  
Included observations: 33 after adjustments  
Maximum dependent lags: 4 (Automatic selection)  
Model selection method: Akaike info criterion (AIC)  
Dynamic regressors (4 lags, automatic): CPS NIM EXRATE INF  
Fixed regressors: C  
Number of models evaluated: 2500  
Selected Model: ARDL(1, 3, 4, 3, 1)  
Variable  
Coefficient  
Std. Error  
t-Statistic  
Prob.*  
GDP(-1)  
CPS  
CPS(-1)  
CPS(-2)  
CPS(-3)  
NIM  
0.918593  
1.981879  
-0.358669  
7.133296  
-8.827085  
1750.944  
2238.716  
248.3360  
1383.260  
-3251.939  
-144.4619  
262.1450  
83.63447  
143.2131  
-128.4055  
172.2031  
-23460.67  
0.081695  
2.195619  
3.321665  
2.928201  
2.203769  
1450.578  
1584.281  
1640.993  
1473.421  
1278.392  
88.11796  
112.4859  
118.5889  
109.5560  
152.0139  
160.9333  
15303.09  
11.24421  
0.902652  
-0.107979  
2.436067  
-4.005450  
1.207067  
1.413080  
0.151333  
0.938809  
-2.543772  
-1.639415  
2.330470  
0.705247  
1.307213  
-0.844696  
1.070028  
-1.533068  
0.0000  
0.3801  
0.9154  
0.0269  
0.0010  
0.2449  
0.1768  
0.8816  
0.3618  
0.0217  
0.1206  
0.0332  
0.4908  
0.2096  
0.4107  
0.3005  
0.1448  
NIM(-1)  
NIM(-2)  
NIM(-3)  
NIM(-4)  
EXRATE  
EXRATE(-1)  
EXRATE(-2)  
EXRATE(-3)  
INF  
INF(-1)  
C
R-squared  
0.999373 Mean dependent var  
0.998746 S.D. dependent var  
9814.671 Akaike info criterion  
1.54E+09 Schwarz criterion  
-338.2042 Hannan-Quinn criter.  
1594.389 Durbin-Watson stat  
0.000000  
286690.9  
277200.5  
21.52753  
22.29846  
21.78692  
2.206306  
Adjusted R-squared  
S.E. of regression  
Sum squared resid  
Log likelihood  
F-statistic  
Prob(F-statistic)  
*Note: p-values and any subsequent tests do not account for model  
selection.  
Page 3673  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
Long-Run Estimation  
ARDL Long Run Form and Bounds Test  
Dependent Variable: D(GDP)  
Selected Model: ARDL(1, 3, 4, 3, 1)  
Case 2: Restricted Constant and No Trend  
Date: 02/19/24 Time: 21:50  
Sample: 1986 2022  
Included observations: 33  
Conditional Error Correction Regression  
Variable  
Coefficient  
Std. Error  
t-Statistic  
Prob.  
C
-23460.67  
-0.081407  
-0.070579  
2369.317  
344.5306  
43.79755  
1.981879  
1.693789  
8.827085  
1750.944  
1620.342  
1868.678  
3251.939  
-144.4619  
-226.8476  
-143.2131  
-128.4055  
15303.09  
0.081695  
1.846356  
2007.776  
70.28809  
148.6213  
2.195619  
2.215558  
2.203769  
1450.578  
1540.099  
1214.529  
1278.392  
88.11796  
114.8336  
109.5560  
152.0139  
-1.533068  
-0.996483  
-0.038226  
1.180070  
4.901693  
0.294692  
0.902652  
0.764498  
4.005450  
1.207067  
1.052102  
1.538603  
2.543772  
-1.639415  
-1.975446  
-1.307213  
-0.844696  
0.1448  
0.3338  
0.9700  
0.2552  
0.0002  
0.7720  
0.3801  
0.4557  
0.0010  
0.2449  
0.3084  
0.1434  
0.0217  
0.1206  
0.0657  
0.2096  
0.4107  
GDP(-1)*  
CPS(-1)  
NIM(-1)  
EXRATE(-1)  
INF(-1)  
D(CPS)  
D(CPS(-1))  
D(CPS(-2))  
D(NIM)  
D(NIM(-1))  
D(NIM(-2))  
D(NIM(-3))  
D(EXRATE)  
D(EXRATE(-1))  
D(EXRATE(-2))  
D(INF)  
* p-value incompatible with t-Bounds distribution.  
Levels Equation  
Case 2: Restricted Constant and No Trend  
Variable  
Coefficient  
Std. Error  
t-Statistic  
Prob.  
CPS  
NIM  
EXRATE  
INF  
C
-0.866980  
29104.47  
4232.181  
538.0049  
-288188.7  
23.48368  
28443.15  
3927.166  
2127.282  
325450.4  
-0.036918  
1.023251  
1.077668  
0.252907  
-0.885507  
0.9710  
0.3214  
0.2972  
0.8036  
0.3890  
EC = GDP - (-0.8670*CPS + 29104.4684*NIM + 4232.1810*EXRATE +  
538.0049*INF - 288188.6825)  
F-Bounds Test  
Test Statistic  
Null Hypothesis: No levels relationship  
Value  
Signif.  
I(0)  
I(1)  
Asymptotic: n=1000  
F-statistic  
k
8.445798  
4
10%  
5%  
2.5%  
1%  
2.2  
2.56  
2.88  
3.29  
3.09  
3.49  
3.87  
4.37  
Actual Sample Size  
33  
Finite Sample: n=35  
2.46  
10%  
5%  
1%  
3.46  
2.947  
4.093  
4.088  
5.532  
Finite Sample: n=30  
2.525  
3.058  
4.28  
10%  
5%  
1%  
3.56  
4.223  
5.84  
Page 3674  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
Short-Run  
Diagnostic Test  
Normality Test  
9
8
7
6
5
4
3
2
1
Series: Residuals  
Sample 1990 2022  
Observations 33  
Mean  
Median  
2.98e-11  
464.4039  
18358.76  
-17091.21  
6940.020  
0.043125  
3.885768  
Maximum  
Minimum  
Std. Dev.  
Skewness  
Kurtosis  
Jarque-Bera 1.089032  
Probability 0.580123  
0
-15000 -10000  
-5000  
0
5000  
10000  
15000  
20000  
Page 3675  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
Model Two  
Method: ARDL  
Date: 02/19/24 Time: 22:25  
Sample (adjusted): 1990 2022  
Included observations: 33 after adjustments  
Maximum dependent lags: 4 (Automatic selection)  
Model selection method: Akaike info criterion (AIC)  
Dynamic regressors (4 lags, automatic): NPI NDI EXRATE INF  
Fixed regressors: C  
Number of models evaluated: 2500  
Selected Model: ARDL(4, 4, 4, 4, 4)  
Variable  
Coefficient  
Std. Error  
t-Statistic  
Prob.*  
GDP(-1)  
GDP(-2)  
GDP(-3)  
GDP(-4)  
NPI  
NPI(-1)  
NPI(-2)  
NPI(-3)  
NPI(-4)  
NDI  
0.738601  
-0.140919  
-0.687098  
0.841945  
-0.091612  
0.019112  
-0.038835  
-0.011469  
-0.021692  
-0.016124  
-0.065703  
-0.008778  
0.094093  
-0.133823  
55.55277  
333.6332  
2.610986  
175.6222  
155.0810  
-199.0215  
280.7106  
-167.9634  
197.3047  
76.76570  
-3260.083  
0.167725  
0.239719  
0.253694  
0.203732  
0.016445  
0.012792  
0.014579  
0.013921  
0.016908  
0.013569  
0.014406  
0.024003  
0.028545  
0.026929  
83.35716  
81.61596  
77.69473  
76.34111  
81.89616  
102.1823  
130.8533  
134.3069  
110.1508  
92.98453  
7382.140  
4.403655  
-0.587850  
-2.708374  
4.132613  
-5.570888  
1.494001  
-2.663772  
-0.823833  
-1.282953  
-1.188256  
-4.560648  
-0.365698  
3.296365  
-4.969475  
0.666443  
4.087843  
0.033606  
2.300492  
1.893629  
-1.947710  
2.145231  
-1.250594  
1.791223  
0.825575  
-0.441618  
0.0023  
0.5728  
0.0267  
0.0033  
0.0005  
0.1735  
0.0286  
0.4339  
0.2354  
0.2688  
0.0018  
0.7241  
0.0109  
0.0011  
0.5239  
0.0035  
0.9740  
0.0504  
0.0949  
0.0873  
0.0643  
0.2464  
0.1110  
0.4330  
0.6705  
NDI(-1)  
NDI(-2)  
NDI(-3)  
NDI(-4)  
EXRATE  
EXRATE(-1)  
EXRATE(-2)  
EXRATE(-3)  
EXRATE(-4)  
INF  
INF(-1)  
INF(-2)  
INF(-3)  
INF(-4)  
C
R-squared  
0.999920 Mean dependent var  
0.999682 S.D. dependent var  
4945.404 Akaike info criterion  
1.96E+08 Schwarz criterion  
-304.1484 Hannan-Quinn criter.  
4188.791 Durbin-Watson stat  
0.000000  
286690.9  
277200.5  
19.94839  
21.08211  
20.32985  
2.063749  
Adjusted R-squared  
S.E. of regression  
Sum squared resid  
Log likelihood  
F-statistic  
Prob(F-statistic)  
*Note: p-values and any subsequent tests do not account for model  
selection.  
Page 3676  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
Long-Run  
Dependent Variable: D(GDP)  
Selected Model: ARDL(4, 4, 4, 4, 4)  
Case 2: Restricted Constant and No Trend  
Date: 02/19/24 Time: 22:27  
Sample: 1986 2022  
Included observations: 33  
Conditional Error Correction Regression  
Variable  
Coefficient  
Std. Error  
t-Statistic  
Prob.  
C
GDP(-1)*  
NPI(-1)  
NDI(-1)  
EXRATE(-1)  
INF(-1)  
D(GDP(-1))  
D(GDP(-2))  
D(GDP(-3))  
D(NPI)  
-3260.083  
-0.247472  
-0.144496  
-0.130334  
722.5001  
187.7960  
-0.013927  
-0.154846  
-0.841945  
-0.091612  
0.071996  
0.033161  
0.021692  
-0.016124  
0.048507  
0.039730  
0.133823  
55.55277  
-333.3141  
-330.7031  
-155.0810  
-199.0215  
-106.1070  
-274.0704  
-76.76570  
7382.140  
0.056334  
0.025544  
0.035110  
104.9756  
169.7986  
0.179890  
0.187353  
0.203732  
0.016445  
0.015816  
0.017758  
0.016908  
0.013569  
0.033001  
0.027972  
0.026929  
83.35716  
105.1907  
92.74182  
81.89616  
102.1823  
152.2080  
102.4690  
92.98453  
-0.441618  
-4.392927  
-5.656854  
-3.712118  
6.882553  
1.105993  
-0.077420  
-0.826496  
-4.132613  
-5.570888  
4.551987  
1.867358  
1.282953  
-1.188256  
1.469858  
1.420349  
4.969475  
0.666443  
-3.168664  
-3.565847  
-1.893629  
-1.947710  
-0.697118  
-2.674667  
-0.825575  
0.6705  
0.0023  
0.0005  
0.0059  
0.0001  
0.3009  
0.9402  
0.4325  
0.0033  
0.0005  
0.0019  
0.0988  
0.2354  
0.2688  
0.1798  
0.1933  
0.0011  
0.5239  
0.0132  
0.0073  
0.0949  
0.0873  
0.5055  
0.0282  
0.4330  
D(NPI(-1))  
D(NPI(-2))  
D(NPI(-3))  
D(NDI)  
D(NDI(-1))  
D(NDI(-2))  
D(NDI(-3))  
D(EXRATE)  
D(EXRATE(-1))  
D(EXRATE(-2))  
D(EXRATE(-3))  
D(INF)  
D(INF(-1))  
D(INF(-2))  
D(INF(-3))  
* p-value incompatible with t-Bounds distribution.  
Levels Equation  
Case 2: Restricted Constant and No Trend  
Variable  
Coefficient  
Std. Error  
t-Statistic  
Prob.  
NPI  
NDI  
EXRATE  
INF  
C
-0.583889  
-0.526661  
2919.527  
758.8588  
-13173.56  
0.113563  
0.078476  
373.7459  
621.2102  
28489.17  
-5.141523  
-6.711075  
7.811528  
1.221581  
-0.462406  
0.0009  
0.0002  
0.0001  
0.2566  
0.6561  
EC = GDP - (-0.5839*NPI -0.5267*NDI + 2919.5267*EXRATE + 758.8588*INF  
- 13173.5618)  
F-Bounds Test  
Test Statistic  
Null Hypothesis: No levels relationship  
Value  
Signif.  
I(0)  
I(1)  
Asymptotic: n=1000  
F-statistic  
k
10.70927  
4
10%  
5%  
2.5%  
1%  
2.2  
2.56  
2.88  
3.29  
3.09  
3.49  
3.87  
4.37  
Actual Sample Size  
33  
Finite Sample: n=35  
2.46  
10%  
5%  
1%  
3.46  
2.947  
4.093  
4.088  
5.532  
Finite Sample: n=30  
2.525  
3.058  
4.28  
10%  
5%  
1%  
3.56  
4.223  
5.84  
Page 3677  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
Error Correction  
Dependent Variable: D(GDP)  
Selected Model: ARDL(4, 4, 4, 4, 4)  
Case 2: Restricted Constant and No Trend  
Date: 02/19/24 Time: 22:29  
Sample: 1986 2022  
Included observations: 33  
ECM Regression  
Case 2: Restricted Constant and No Trend  
Variable  
Coefficient  
Std. Error  
t-Statistic  
Prob.  
D(GDP(-1))  
D(GDP(-2))  
D(GDP(-3))  
D(NPI)  
D(NPI(-1))  
D(NPI(-2))  
D(NPI(-3))  
D(NDI)  
-0.013927  
-0.154846  
-0.841945  
-0.091612  
0.071996  
0.033161  
0.021692  
-0.016124  
0.048507  
0.039730  
0.133823  
55.55277  
-333.3141  
-330.7031  
-155.0810  
-199.0215  
-106.1070  
-274.0704  
-76.76570  
-0.247472  
0.102822  
0.125238  
0.127750  
0.010196  
0.008726  
0.009821  
0.009744  
0.008013  
0.015141  
0.014266  
0.017127  
44.54259  
69.30068  
61.57887  
51.42325  
62.28771  
62.78011  
58.78612  
52.94694  
0.024218  
-0.135448  
-1.236412  
-6.590578  
-8.985434  
8.251156  
3.376459  
2.226172  
-2.012099  
3.203804  
2.785007  
7.813651  
1.247183  
-4.809680  
-5.370399  
-3.015776  
-3.195197  
-1.690137  
-4.662161  
-1.449861  
-10.21838  
0.8956  
0.2514  
0.0002  
0.0000  
0.0000  
0.0097  
0.0566  
0.0790  
0.0125  
0.0237  
0.0001  
0.2476  
0.0013  
0.0007  
0.0167  
0.0127  
0.1295  
0.0016  
0.1851  
0.0000  
D(NDI(-1))  
D(NDI(-2))  
D(NDI(-3))  
D(EXRATE)  
D(EXRATE(-1))  
D(EXRATE(-2))  
D(EXRATE(-3))  
D(INF)  
D(INF(-1))  
D(INF(-2))  
D(INF(-3))  
CointEq(-1)*  
R-squared  
0.990467 Mean dependent var  
0.976533 S.D. dependent var  
3879.494 Akaike info criterion  
1.96E+08 Schwarz criterion  
-304.1484 Hannan-Quinn criter.  
2.063749  
27924.48  
25325.06  
19.64536  
20.55233  
19.95053  
Adjusted R-squared  
S.E. of regression  
Sum squared resid  
Log likelihood  
Durbin-Watson stat  
* p-value incompatible with t-Bounds distribution.  
F-Bounds Test  
Test Statistic  
Null Hypothesis: No levels relationship  
Value  
Signif.  
I(0)  
I(1)  
F-statistic  
k
10.70927  
4
10%  
5%  
2.5%  
1%  
2.2  
2.56  
2.88  
3.29  
3.09  
3.49  
3.87  
4.37  
Diagnostic Test  
Normality Test  
7
6
5
4
3
2
1
0
Series: Residuals  
Sample 1990 2022  
Observations 33  
Mean  
-1.10e-10  
Median  
63.05613  
5405.563  
-6396.285  
2472.702  
-0.318391  
3.321330  
Maximum  
Minimum  
Std. Dev.  
Skewness  
Kurtosis  
Jarque-Bera 0.699525  
Probability 0.704856  
-6000  
-4000  
-2000  
0
2000  
4000  
6000  
Page 3678  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
Null hypothesis: No serial correlation at up to 2 lags  
F-statistic  
Obs*R-squared  
0.189722 Prob. F(2,6)  
1.962815 Prob. Chi-Square(2)  
0.8320  
0.3748  
Test Equation:  
Dependent Variable: RESID  
Method: ARDL  
Date: 02/19/24 Time: 22:31  
Sample: 1990 2022  
Included observations: 33  
Presample missing value lagged residuals set to zero.  
Variable  
Coefficient  
Std. Error  
t-Statistic  
Prob.  
GDP(-1)  
GDP(-2)  
GDP(-3)  
GDP(-4)  
NPI  
NPI(-1)  
NPI(-2)  
NPI(-3)  
NPI(-4)  
NDI  
0.148161  
-0.107302  
-0.158045  
0.092765  
-0.004958  
0.003100  
-0.007188  
0.003537  
0.005753  
-0.004984  
-0.007006  
0.002192  
-0.001892  
-0.008816  
57.84407  
16.61317  
-41.00254  
0.155850  
-40.35571  
-90.99118  
92.73509  
-69.17627  
58.46489  
-21.69310  
549.3479  
-0.324641  
-0.594424  
0.306888  
0.351806  
0.399995  
0.274156  
0.020671  
0.015194  
0.021148  
0.017440  
0.021879  
0.020715  
0.020011  
0.027263  
0.032650  
0.034167  
136.5466  
95.37549  
109.6440  
86.05997  
114.3949  
187.7173  
210.1447  
187.7020  
158.1236  
109.9220  
8499.888  
0.718288  
1.015132  
0.482786  
-0.305003  
-0.395118  
0.338366  
-0.239879  
0.204041  
-0.339877  
0.202781  
0.262958  
-0.240585  
-0.350131  
0.080404  
-0.057953  
-0.258032  
0.423622  
0.174187  
-0.373961  
0.001811  
-0.352776  
-0.484725  
0.441292  
-0.368543  
0.369742  
-0.197350  
0.064630  
-0.451965  
-0.585564  
0.6464  
0.7707  
0.7064  
0.7466  
0.8184  
0.8451  
0.7455  
0.8460  
0.8014  
0.8179  
0.7382  
0.9385  
0.9557  
0.8050  
0.6866  
0.8674  
0.7213  
0.9986  
0.7363  
0.6451  
0.6745  
0.7251  
0.7243  
0.8501  
0.9506  
0.6672  
0.5795  
NDI(-1)  
NDI(-2)  
NDI(-3)  
NDI(-4)  
EXRATE  
EXRATE(-1)  
EXRATE(-2)  
EXRATE(-3)  
EXRATE(-4)  
INF  
INF(-1)  
INF(-2)  
INF(-3)  
INF(-4)  
C
RESID(-1)  
RESID(-2)  
R-squared  
0.059479 Mean dependent var  
-4.016111 S.D. dependent var  
5538.030 Akaike info criterion  
1.84E+08 Schwarz criterion  
-303.1366 Hannan-Quinn criter.  
0.014594 Durbin-Watson stat  
1.000000  
-1.10E-10  
2472.702  
20.00828  
21.23270  
20.42026  
1.851818  
Adjusted R-squared  
S.E. of regression  
Sum squared resid  
Log likelihood  
F-statistic  
Prob(F-statistic)  
Page 3679  
www.rsisinternational.org