eco 309 pp3

ECO 309

Economic Forecasting

Project Part 3

Regression Analysis

Data Prep: Make sure you have 2-3 related X (macroeconomic) variables. Remember that these variables will be quarterly as your Y (Revenue) variable. All variables should be from the same time period. Recall that the first column is dates and second column is your Y variable. Your third column will be X1, 4th will be X2 and so forth. Label your X variables properly so I can understand what those variables are.

1. Create a correlation matrix with p-values (you can find it under Stat>Basic Statistics>Correlation) of Y and all Xs. You need to tell me which Xs are more related to Y and which Xs are not. Also check the correlation among the X variables. Write a paragraph about possible negative or positive relation between each X with Y and the X variables among themselves based on the direction of the relationships and the statistical significance based on. The p-value. In addition, you are welcome to use the scatter plots to make your point. (watch for assumption violation on non-linearity and/or multicollinearity?)

2. Run your first regression and have MINITAB report the VIF and DW test statistic (Stats>Regression>Regression. You can find VIF and DW under options.)

3. Explain your results. Your explanation should include the following paragraphs.

(i) Interpret regression equation; does it make sense to you?

(ii) Which variables are statistically significant? Which are not?

(iii) Which variables are highly correlated?

(iv) Do you have a “sign switch”?

(v) Do you have nonlinear relationship between Y & and of the X variables?

(vi) Do you have outliers? Are you planning to do anything about them?

(vii) Which of the regression assumptions is your model violating?

(viii) Do you have a good overall model? Use at least 2 ways to illustrate.