Multiple regression is use to evaluate the relationship between a response variable and more than
one explanatory variable. The issue of selecting the best set of regressors to be included in the model that best
explain the response variable was handled by all possible regression variable selection technique. Ridge regression
was also applied in this research work, as means of dealing the effect of multicolinearity on the model estimate. The
model obtained when the effect of multicolinearity was neglected before the variable selection process was
compared with the model gotten when the effect has been remedied. And we concluded that the model obtained when
the effect of multicolinearity has been dealt with before the variable selection process by ridge regression was better
than the model when the effect of multicolinearity has been neglected (ordinary least square method) because it is
both precise and accurate.

Download Full Page



<<< Go Back