Graded Lab
- In this lab you will work with Cereal Ratings Dataset, same as last week
- You will work with SAP and QlikView and document results in
this Word
File: MultipleLinearSummary.docx
. Make sure you
complete CONCLUSION part in the Word File
- Open SAP and open file we worked last week. Check you have
cerealsTrainSetNov23 dataset loaded.
- Run Multiple Linear Regression Models listed in the Word file. There are 3 new models listes. To receive a
full
grade you must complete at least TWO.
- Document R^2, Adjusted R^2 and Multiple Linear Regression Formula
for each Model in the Word file MultipleLinearSummary2018.docx
- Rate the models based on R^2 and adjusted R^2. Assume, 1 is a HIGHEST RATING
- Open QlikView Test file we worked on Friday. If you missed the class, download the QlikView test file and save it on your
computer with your initials at the end of the file name.
- Create a new sheet, call it Lab9.
- Create a Straight Table Chart to document results for new models (Dimension
is name). Calculate Predicted rating for EACH model and SSE for Each
model.
IMPORTANT!!!For ALL expressions on Number Tab select Fixed to 2 Decimals
- Create a Combo Chart to display predicted rating vs actual rating for new models.
- Find Test_R^2 = 1-SSE/SST for each model on the Test Data
- Document SSE and Test_R^2 for each model in the Word File
- Complete Correlation Table in the Word File. To calculate correlations, add Text Object and use the following
formula:
=Num(Correl(Variable1, Variable2))
Complete CONCLUSION part in the Word File
Submit TWO FILES, QlikView file and Word Summary
File
If time permits
Examine additional models. While choosing which variables to include in the model, take in account correlations you
found earlier.