Qualitalive CHAPTER 7 Multiple Regression Analysis with of the data. Heteroskeda
ID: 3317983 • Letter: Q
Question
Qualitalive CHAPTER 7 Multiple Regression Analysis with of the data. Heteroskedasticity c chapter, th of the partial effects of the explanatory variables for the middle ranges invalidate the usual OLS standard errors and test statistics, but, as we will see in the next easily fixed in large enough samples Section 7.6 provides a discussion of how binary variables are used to evaluate policies and progr As in all regression analysis, we must remember that program participation, or some other binary regre with policy implications, might be correlated with unobserved factors that affect the dependent varia resulting in the usual omitted variables bias. We ended this chapter with a general discussion of how to interpret regression equations when dependent variable is discrete. The key is to remember that the coefficients can be interpreted as the eff on the expected value of the dependent variable. Key Terms Base Group Benchmark Group Binary Variable Chow Statistic Control Group Difference in Slopes Dummy Variable Trap Dummy Variables Experimental Group Interaction Term Intercept Shift Linear Probability Model (LPM) Ordinal Variable Percent Correctly Predicted Policy Analysis Program Evaluation Response Probability Self-Selection Treatment Group Uncentered R-Squared Zero-One Variable Problems 1 Using the data in SLEEP75 (see also Problem 3 in Chapter 3), we obtain the estimated equation sleep 3.840.83-163tonerk-1 1.71 educ-8.70 age (235.11) (018) (5.86) (11.21) + .128 age+87.75 male (134) (34.33) n=706, R' = .123,F-.117. The variable sleep is total minutes per week spent sleeping at night, totwrk is total weekly minut spent working, educ and age are measured in years, and male is a gender dummy (0) All other factors being equal, is there evidence that men sleep more than women? How strong (i) Is there a statistically significant tradeoff between working and sleeping? What is the estimate (ii) What other regression do you need to run to test the null hypothesis that, holding other factors is the evidence? tradeoff? fixed, age has no effect on sleeping? 2 The following equations were estimated using the data in BWGHT: log(hight) = 4.66-.0044 cigs +0093 log(famine) +M6 parity (22) (.0009) (0059) +027 male + 055 white (.006) 010) (.013) " = 1.388, R-0472Explanation / Answer
Answer:
i. The first equation estimates indicates that an increase in cigraettes smoked by 10 units is expected to decrease birth weight by 4.4%, holding all other factors constant.
ii. The first equation indicates that a white child expected to weight 5.5% more than a non-white child, holding all other factors constant. To find if the difference isstatistically significant, we need to compute the t-statistic on white: .055/.013=4.23.For an alpha=5%, and two-sided alternative, with n-k-1=1,388-5-1=1,382 degrees offreedom, the critical value in Table G.2 is c=1.960. Our rejection/fail to reject rule is:Fail to reject null if |t|<cand Reject null if |t|>cSince |4.23|>1.96, we reject the null of insignificance. That is, the difference betweenwhite and nonwhite is statistically different
iii. The coefficient on mothedu indicates one year of mothers education is expected to decrease by birth weight of 30%, holding all other factors constant.
iv. We cannot compute F-statistic because two regressions are different. We would have to re-estimate the first equation using just only the 1,191observations that are in the second regression, and get the R^2.
predicted to decrease birth weight by .30%, holding all other factors constant
The coefficient on motheduc suggests one more year of mother’s education ispredicted to decrease birth weight by .30%, holding all other factors constant