Modelling Nigeria Inflation rate Volatility with Autoregressive Conditional Heteroscedasticity (ARCH) Models
- November 30, 2020
- Posted by: RSIS Team
- Categories: IJRIAS, Mathematics
International Journal of Research and Innovation in Applied Science (IJRIAS) | Volume IV, Issue X, October 2020 | ISSN 2454–6186
Wiri, Leneenadogo & Sibeate,Pius U.
Department of Mathematics,
Rivers State University, Nigeria
Abstracts
This study applied Autoregressive conditional heteroscedasticity (ARCH) models in modelling Nigeria inflation rate. The time plot of the original series showed the present of seasonality and logarithm transformation of return series make it stationary. The return was estimated using both the conditional mean and conditional variance. The study applied both symmetric and asymmetric (GARCH) model that capture the feature of a financial series, such as volatility clustering and leverage effect in modelling the return series of inflation. However, four models were estimated for the conditional mean and seven models were estimated for the conditional variance and asymmetric power autoregressive conditional heteroscedasticity (APARCH (1,1)) was adopted as the best model for the return series and for the conditional mean follow an ARMA (1,1). Finally, the most adequate model for estimating volatility of the inflation rates is the asymmetric APARCH (1.1) model.
Keywords: Inflation rate, conditional means, conditional variance, GARCH, Volatility clustering.
1.0 Introduction
The inflationary period is a time of increase in price of goods and services. This period, the quantity and type of products (good and services) purchase by individuals and corporate body at any point in time is low. The problem cause individuals and corporate bodies in the societies to purchase below the deserved quantity of products. During inflation, income earners especially those with poor income and fixed income in the society find it difficult to match with the increasing prices of goods and services. This continues as long as there is increase in price and there is fall in the purchasing power. More values of money is being required by individuals for the purpose of purchasing desired products during the period of inflation as opposed to normal economic situations. [18].
Inflation series are usually very volatile in nature and this pattern can better be model using a stochastics non-linear modelling method that characterise thebehave of the series. Volatility is one of the most important concepts in finance, its measured variances of variable asset return [14]. Volatility is often used as a basic measure of the total risk of financial assets. Policy maker are interested in measuring volatility process to learn about financial market expectation and uncertainty. [8]. A number of models have been developed to investigate volatility across different countries. The most common model to estimate volatility is the GARCH model developed independently by [16]. GARCH model, this modelislinear in mean, but non-linear in variance. The GARCH models is used to estimate the conditional mean and conditional variance. Engle described the conditional variance by a simple quadratic function of its past lagged terms. [4] extended the basic ARCH mode and describes the conditional variances by its own lagged value and the square of the lagged terms of the shock. The GARCH model provide a good technique for analysing financial time series and estimating conditional variance. [6].
