2. (14 points) A friend of yours run the following regression: y-a + zi t + Mn.

Question

2. (14 points) A friend of yours run the following regression: y-a + zi t + Mn. + et. Then she got the ues of residuals, res2, and run another regression. Please find below the Stata output of this last squared val regression gen x12 x1 x2 reg res2 xi x2 x12 Source df MS Number of obs50 P3 46)2.98 Prob > 0.0409 R-squared0.1628 Adj R-squared- 0.1082 Root MSE 6.980 Model 722-651799 3 240.883933 Residual 3716.03673 46 80,7834071 Total 4438.68853 49 90.5854801 res2 coef.Std. Err. t p>It! [95, conf. Interval 1.736243 0.85 0.401 4.96748 2.02227 1.547959 6583362 .094553 ,0675772 1.400.168-,0414729 .2305789 cons 7.39264 13.83216 0.53 0.596 -20.45006 35.23534 x1 1.472605 x24448113 5480401 0.81 0.4211 x12 a) What do you think is the purpose of your friend in running the last regression? b) What can you conclude by looking at the Stata output? Be specific about which values of the output table to check?

Explanation / Answer

In first part, she regressed Y on X1 and X2 only. So in the second part, she was trying to see that weather there is is any effect of interaction term of X1*X2 on the model. (a) By the value of R-squared and adjusted R-squared we can clearly see that model is not a good fit since they are very low. (b) Looking up at the p-value for regression coefficient of x1 and x2 we can say that they are not significant for ou model, since thay are very high. (c) X1*X2 has the highest significance among all chosen independent variabl.

Related Questions

Navigate

• Browse (All)
• Browse 2
• Subjects

---

---

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.