e-Statistics

## Residual Analysis

The standard assumption of regression model is that the random error terms are iid normally distributed random variables with mean 0 and common variance . Thus, the statistical model for the simple linear regression is given by

for .

The data set consists of

1. explanatory variable for 's;
2. dependent variable for 's.
For each value of the independent variable, the prediction equation provides a fitted value . Then we can plot the fitted value against the standardized residuals .

In model validation we look for a pattern, the indication of which suggests that the regression of choice is not a good model.