# Calculating residuals standard error and f-statistics

The residuals, or errors, are the unexplained variations of the model. It is the separation of the expected value to the response given by the linear regression.

f-statistics is another test for rejecting the possibility of the slope being equal to zero.

To calculate the residuals, we will write down the equation for a multivariable regression model. Then, we will calculate how well the model fits the expected values. The difference between the expected values (sales revenue) and the results from the regression model is the errors or unexplained variation. Remember that this variation is supposed to be small, with a small standard deviation to have a model useful for prediction.

f-statistics come from the explained and unexplained variations of the expected values and the linear model. We will calculate the f-statistics and then check whether the value is in the region of rejection for accepting the alternate hypothesis...