Professional Documents
Culture Documents
Screenshot 2023-09-07 at 22.35.34
Screenshot 2023-09-07 at 22.35.34
Screenshot 2023-09-07 at 22.35.34
METHODOLOGY
3.1: Introduction
This section of the study focuses on investigating the impact of inflation on the standard of
living in Nigeria. It outlines and discusses the econometric methodology to be employed, the
types of data to be utilized, the model specification, the data analysis techniques, and a
description of the variables to be incorporated into the models.
3.6: Post-Estimation.
3.6.1: Serial-correlation Test.
The Serial correlation test would be performed to see whether the error corresponding to the
different observation is not correlated and also to ascertain if there is linear correlation between
any two set of the variables, when all other variables are held constant. The Breush-Goffery (BG)
test for autocorrelation will be used test for serial correlation in this study. The level of
significance used is 5 percent.
3.6.2: Heteroscedasticity Test.
From the word heteroskedasticity, ‘Hetero’ means ‘inconstant’ while ‘Skedasticity’ means spread
of variance. Thus, Heteroskedasticity means inconstant spread of variance. An important
assumption of the linear classical regression model is that the disturbance term in a regression
equation is homoscedastic. In line with this therefore, the presence of heteroskedasticity will
Bess tested.
3.6.3: Test for Normality.
Non-normality of error term poses problem of efficiency to the coefficients generated and also
bias of standard error. The normality assumption is that the error term must be identically
independently distributed (I.I.D), thus to test for the satisfaction of this assumption the Jaque-
Bera (JB) test for normality will be carried out.