A problem of interest to health officials (and others) is to determine the effects of smoking during pregnancy on infant health. One measure of infant health is birth weight; a birth weight that is too low can put an infant at risk for contracting various illnesses. Since factors other than cigarette smoking that affect birth weight are likely to be correlated with smoking, we should take those factors into account. For example, higher income generally results in access to better prenatal care, as well as better nutrition for the mother. An equation that recognizes this can be

bwght = beta0 + beta1*cigs + beta2*famincome + betha3*white + beta4*male + u

where bwght is the weight of the baby, cigs is the number of cigarettes smoked per day, famincome is the household income, white a dummy variable for race, and male a dummy variable for gender.

i. What are the most likely signs for beta1, beta2, beta3, and beta4? Why

ii. Do you think cigs and faminc are likely to be correlated? Explain why the correlation might be positive or negative.

iii. Now, estimate the equation using Eviews with and without white and male, using the data in BWGHT file. Report the results in equation form, including the sample size and R-squared. Interpret the value of R-squared, and the value of the estimated coefficients. Are the estimated coefficients statistically significant? Explain. Discuss your results, focusing on whether adding white and male substantially changes the estimated effect of cigs on bwght.

iv. Construct a 95% confidence interval for beta1. Interpret the confidence interval.

v. Perform a joint hypothesis test on whether beta3=beta4=0. What do you conclude? Is there evidence to reject the null hypothesis?