Quality Kitchen

QUALITY KTICHEN

INTRODUCTION: The aim of this exercise is to provide an analysis that aims to quantify the effect of several variables into the sales patterns of Amalgamated Food Products meat loaf brand. In doing so, we intend to be able to produce a multiple regression forecasting model that will allow us to analyze the relationships between these factors and the sales outcomes. We started our process by looking into the individual variables that were provided, namely, the expenditures for promotion and advertising, an index of economic conditions, as well as seasonality patterns. Soon enough, we realized that looking into each of them individually was not yielding the desired approach and instead we moved to interpreting the information provided and combining all of it into a holistic approach. In the end, the variables that were selected and their rationale are highlighted below: Promotion: amount spent on promotions; Promo_lag: amount spent on promotions during the previous month, that is for sales at t=2 we looked at the amount spent on promotions on t=1; Advertising: amount spent on advertising; Advert_lag: amount spent on advertising during the previous month, calculated in the same way as Promo_lag is described above; Seasonality: entered as a dummy variable and measured between winter (Q1 & Q4) and summer (Q2 & Q3); Econ Index de-trended: the original economic index appeared to have an upward sloping trend, therefore we attempted to reduce this by taking a difference from average approach, calculated as each value minus the average of all observations; 6m % change in Econ Index: the percentage change between two quarters.

-

(For the full table of values please refer to Exhibit 1 in Appendix) Afterwards and in order to avoid the potential pitfalls of multi-colinearity, we started by testing the correlation between the variables described above. Exhibit 2 in Appendix shows the correlation table and it illustrates that none of the chosen...