/* Make table 2 for duration paper */ /* AS NOTED MORE THAN ONCE IN THE PAPER, these duration models are parameterized so that the coefficients estimate the multiplicative effect of a one unit change in an explanatory variable. In Stata this is referred to as the "accelerated failure time" metric. Thus the coefficients are slightly different than the usual hazard rate parameterization. You can look it up! */ streg couprev eeur nocon sos drugs ,d(w) nolog time tr streg couprev eeur nocon sos drugs emELF ,d(w) nolog time tr streg couprev eeur nocon sos drugs emgdpl if ~(ccode==200 & year==1969) ,d(w) nolog time tr streg couprev eeur nocon sos drugs lempopl ,d(w) nolog time tr streg couprev eeur nocon sos drugs ethwar ,d(w) nolog time tr streg couprev eeur nocon sos drugs polity2l if anticol~=1,d(w) nolog time tr streg couprev eeur nocon sos drugs linten ,d(w) nolog time tr