This document introduces intervention analysis as it applies to regression models. It describes a case scenario where a student and instructor work through building a regression model to explain and forecast monthly sales (FRED.SALE) for Fred's Deli using monthly advertising (FRED.ADVERT) as a predictor variable. Initially, plotting the sales data over time shows no clear pattern. Looking at autocorrelations reveals some seasonality in sales at 12 months but values are not significant. Cross-correlating sales with advertising shows a significant correlation at a 2 month lag, so a regression model is estimated with sales as the dependent variable and advertising lagged by 2 months as the explanatory variable. Diagnostic checks of the residuals from this model show