Drug companies are not forced to divulge all studies they performed to the FDA. Suppose a drug company knows that the drug has no effect and followed the strategy described in (b1), but the study comes back insignificant. Should the drug company try again, i.e. perform another study?
Suppose the drug company has the policy of conducting at most three studies. If the drug has no effect, what is the distribution of the profits arising from this policy? Compute the expectation of that distribution.
In the files yields.xls and yields.dta, you find the monthly stockmarket returns in percent in the US and Germany from 1950:2 to 2006:4.
Consider a portfolio that consists of 30% German stocks and 70% US stocks (and these proportions remain constant through time). Compute the yield in each month of this portfolio. Is there evidence in the data that this portfolio does significantly better in January compared to all other months? (Finance people call this the January effect...) Conduct an appropriate one-sided hypothesis test on the 5% level.