Archive | 2019
Linear Regression Model: Relaxing the Classical Assumptions
Abstract
This chapter relaxes the homoscedasticity and nonautocorrelation assumptions of the random error of a linear regression model and shows how the parameters of the linear model are correctly estimated and tested in presence of heteroscedastic and autocorrelated error in the model. Random errors are heteroscedastic when they have different variances for different predictors. Heteroscedasticity is a problem mainly for cross section data. The problem of autocorrelation arises when errors are serially correlated. This problem is usually found in time series data. In time series, autocorrelation is the correlation of a variable with lags of itself. Presence of autocorrelation implies that current error can remember its past values.