Following table (download in Excel here Download Excel here) shows 100 individual data on health status (score between 0-100), weekly exercise hours, annual income, and diet score (between 0-100).
The goal is to explore the relationship between these variables, so that you can predict health score based on all or some of other variables. To achieve this goal:
Create scatter plots of health score with each of other variables and interpret the patterns briefly.
Analyze the correlation table among the variables. Interpret the correlation matrix.
Develop a regression model using all variables and estimate the coefficients using Excel (or other software). Interpret the results, including R square, adjusted R square, and p-values.
Based on your conclusion in previous step, develop a new regression model (by removing some of insignificant variables) and compare the R square and p-values of the new model with old model.
