consider taking a large random sample of workers at a given point in time. let if person called in sick during the last 90 days, and zero otherwise. let be a vector of individual and employer characteristics. let be the number of cigarettes individual smokes per day (on average). (i) explain the underlying experiment of interest when we want to examine the effects of cigarette smoking on workdays lost; (ii) why might be correlated with unobservable affecting ?; (iii) one way to write the model of interest is where is a subset of and is an unobservable variable that is possibly correlated with . what happens if is ignored and you estimate the probit of on , ? (iv) can have a conditional normal distribution in the population? explain. (v) explain how to test whether is exogenous. does this test rely on having a conditional normal distribution? (vi) suppose some of the workers live in states that recently implemented no-smoking laws in the workplace. does the presence of the new laws suggest a good iv candidate for ?