Homework 3

Using the project data on Table spreads, please do the following.

1. Use 2018 data for table spreads (one year's data). Create a weekly market share table for different brands in the table spread category (e.g., Parkay is a brand. Combine all UPCs of Parkay into one brand). Market share = brand sales ($)/ total category sales ($)
2. Do the above for only the top 5 brands in each category (i.e., butter and margarine to be done separately) and keep all other UPCs in one brand call it "Other". Do this analysis for all of US only.
3. Compute average price in $ per 16 ounces for each brand in each week.
4. Compute average price decrease when the brand is on Merchandising (say 5% or 10%).
5. Create a dummy variable to indicate a week when there is a merchandising for a brand (here if any UPC of a brand is on promotion, the whole brand is considered to be on promotion and takes the value =1, else it is 0). This will be useful if you wish to compare frequency of merchandising across brands.

Brands	Week	Market share	Average price	Average promotion price decrease	Whether on promo or not (1 or a 0)
1	1
1	2
1	3
1	4
1	...
2	1
2	2
2	3
2	4

1. Run a OLS regression model (for each of 4 Conagra brands shown in the powerpoint) to predict Market share using the above variables or any other variables that you think can be included in the model. Write a careful report on your findings. Calculate price elasticity for Conagra brands.
2. Now run a Random intercept model and a Fixed Intercept model (RANONE and FIXONE) using Brand as a cross section and week as a time variable. Use PROC PANEL (use the same independent variables as above). Comment on your new findings and calculate price elasticity for Conagra Brands.

i want sas code