Autocorrelation Determination in Several Regression Models’ Residual With Nigeria Gross Domestic Product (Gdp) Data

Six regression models have been utilized in this particular research to check for the existence of correlation in model residual using secondary data on Nigeria Gross Domestic Product at Current Basic Prices.The appropriate model for this investigation is the Sine design that has the very least mean square error, the very least standard deviation, smallest AIC, smallest BIC and HQC, although the particular design contains the greatest R 2 valued while the Cosine design has the greatest Durbin Watson statistical value, but many parameters came up because of transformation effect.It was concluded from this study that the sine transformation technique could be more effective in solving the issue of autocorrelation in real life data.Therefore, it was suggested that much more comprehensive work be performed making use of other transformation methods not mentioned or even used in this particular current study to determine a far more appropriate design to be used in the research or for dealing with autocorrelation issues in the data.