E9.2 Use the data set Birthweight_Smoking introduced in Empirical Exercise 5.1 to answer the following questions.

a. In Empirical Exercise 7.1(f), you estimated several regressions and were asked: “What is a reasonable 95% confidence interval for the effect of smoking on birth weight?”

i. In Chapter 8 you learned about nonlinear regressions. Can you think of any nonlinear regressions that can potentially improve your answer to Empirical Exercise E7.1(f)? After estimating these additional regressions, what is a reasonable 95% confidence interval for the effect of smoking on birth weight?

ii. Discuss the internal validity of the regressions you used to construct the confidence interval. Include a discussion of possible omitted variable bias, misspecification of the functional form of the regression, errors in variables, sample selection, simultaneous causality, and inconsistency of the OLS standard errors.

b. The data set Birthweight_Smoking includes babies born in Pennsylvania in 1989. Discuss the external validity of your analysis for (i) California in 1989, (ii) Illinois in 2015, and (iii) South Korea in 2014.