Get fast, custom help from our academic experts, any time of day.
Place your order now for a similar assignment and have exceptional work written by our team of experts.
Secure Original On Schedule
At 5.8 deaths per 1,000 live births, the United States ranks No. 33 out of 36 OECD countries: infant mortality is not low here, and the reasons why are not well understood. Low birth weight seems to be very important however: about 3/4 of infant deaths can be attributed to infants being born too small or too early. But a baby’s birth weight depends on many complicated interactions between factors that are still not well understood by medicine. But maybe econometrics can help…
Suppose we wanted to figure out the effect of maternal smoking on birth weight:
1) (4pts) Consider these 3 variables: multiple_births (whether or not the mother is pregnant with twins, triplets, etc), drinking (whether or not the mother drinks alcohol when pregnant), and prenatal_vitamins (whether or not the mother takes prenatal vitamins). Assume we have to omit these variables from the model above because we don’t have data about them. Which would cause omitted variable bias, and which direction would they bias the estimate for ? That is, would omitting any of them make it look like smoking is more or less harmful to the baby? Explain your answers.
2) (4pts) How would you design the ideal experiment to evaluate the effect of maternal smoking on birth weight?
3) (4pts) Consider these two potential instruments: state cigarette taxes (the tax per pack of cigarettes charged by the state the mother lives in) and premature birth (how many weeks early the baby is born, if it is born early). Argue why/why not each of these instruments would be valid, going into detail about each of the three conditions for a valid instrument.
4) (4pts) Suppose we had a valid instrument z and we ran 2SLS, and got these results:
What does refer to? What is the estimate for the causal effect of x on y?
Get fast, custom help from our academic experts, any time of day.
Secure Original On Schedule