{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

Final Exam Econometrics

# Final Exam Econometrics - Kate Brown Dr Cuellar...

This preview shows pages 1–7. Sign up to view the full content.

Kate Brown Dr. Cuellar Econometrics Final a). reg wage exper exper2 educ looks female Source | SS df MS Number of obs = 1260 -------------+------------------------------ F( 5, 1254) = 64.15 Model | 5569.95985 5 1113.99197 Prob > F = 0.0000 Residual | 21777.4793 1254 17.3664109 R-squared = 0.2037 -------------+------------------------------ Adj R-squared = 0.2005 Total | 27347.4392 1259 21.7215561 Root MSE = 4.1673 ------------------------------------------------------------------------------ wage | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- exper | .257418 .0384844 6.69 0.000 .181917 .332919 exper2 | -.0039446 .0008609 -4.58 0.000 -.0056335 -.0022557 educ | .4095647 .0464471 8.82 0.000 .3184421 .5006872 looks | -.4049482 .1751361 -2.31 0.021 -.7485403 -.0613562 female | -2.503287 .255191 -9.81 0.000 -3.003935 -2.002639 _cons | .3519805 .8714727 0.40 0.686 -1.357725 2.061686 ------------------------------------------------------------------------------ b) Overall significance of the model : Adj R-squared = 0.2005: meaning that 20.05% of the regression is explained by the model. exper | .257418 t-stat 6.69 exper2 | -.0039446 t-stat -4.58 educ | .4095647 t-stat 8.82 looks | -.4049482 t-stat -2.31 female | -2.503287 t-stat -9.81 At first it seems that looks have a negative affect on wage. . display invFtail(5,1258,.05) 2.2212133 F( 5, 1254) = 64.15

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
c) 0 5 10 15 20 Standardized residuals 0 2 4 6 8 10 Fitted values The graph above compares the Standardized residuals to the fitted values. It would appear there is a
negative trend. The graph is heteroskedastic. I would assume that in the lower income jobs that the effect of beauty may not have much of an impact. d) Yes, there is an obvious outlier at 4.5,18 as seen in the graphed residual analysis below: 0 20 40 60 80 Residuals 0 2 4 6 8 10 Fitted values e) Cook’s Distance measures how much one particular observation alone affects the regression estimates. After a regression is run, you perform a Cook’s Distance test to recognize how far away that point is from the means of the independent variables and dependent variable. In our case we observe influential points as outliers in the Hammer mesh and Biddle data. Cook’s Distance: 4/n : Deletes one Observation at a time and measures the change in slope. display 4/1260 .0031746 Observations seen with a change in slope over .0031746 impact our model and could be problematic.

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
f) . list looks wagehat D if D> .003 +-----------------------------+ | looks wagehat D | |-----------------------------| 57. | 3 4.788664 .0030575 | 267. | 4 4.436476 .0120295 | 337. | 3 5.230033 .009749 | 348. | 2 4.360388 .1786339 | 404. | 4 6.236249 .0044739 | |-----------------------------| 431. | 3 8.454038 .0066791 | 474. | 3 8.62073 .0173239 | 524. | 3 5.912581 .0110585 | 615. | 2 7.840163 .0036605 | 627. | 1 9.883369 .0149926 | |-----------------------------| 649. | 2 7.975298 .0329545 | 679. | 1 10.14575 .0068272 | 686. | 3 9.335855 .0112907 | 862. | 3 7.957862 .004132 | 863. | 3 7.957862 .0072708 | |-----------------------------| 881. | 3 5.090909 .0033359 | 909. | 3 7.202866 .0055904 | 966. | 3 9.805318 .0088153 | 971. | 3 10.21488 .0044642 | 988. | 3 10.24746 .0035539 | |-----------------------------| 996. | 3 8.6092 .0185368 | 1041. | 3 8.659595 .0106435 | 1043. | 1 8.194899 .0043905 | 1064. | 2 10.29426 .025588 | 1069. | 2 7.010871 .0117475 | |-----------------------------| 1071. | 3 8.244181 .0039533 | 1094. | 3 10.27725 .0045533 | 1095. | 3 8.229423 .0039603 | 1098. | 2 10.68219 .003071 | 1114. | 3 8.206775 .0073861 | |-----------------------------| 1134. | 3 8.176238 .0100178 | 1146. | 4 7.77129 .0168406 | 1179. | 4 9.680167 .0145949 | 1197. | 4 6.813215 .0036984 | 1237. | 3 9.379842 .017562 | |-----------------------------| 1257. | 2 2.471673 .0034372 | +-----------------------------+ This is the observation of the obvious potentially problematic outlier. 348. | 2 4.360388 .1786339 |
The Cooks test shows that the outlier has a significant impact on the slope. Weighted with Cook’s Test: 0 5 10 15 20 Standardized residuals 0 2 4 6 8 10 Fitted values We find that the observations with larger bubbles are the ones with the most impact on our model.

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
4 2 3 3 2 3 3 2 5 2 3 3 3 3 3 4 2 2 4 3 3 4 2 3 2 2 3 3 3 2 3 3 2 4 3 2 3 2 3 3 2 3 3 3 3 3 3 3 3 3 3 3 3 2 2 2 3 3
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}