I am a bit puzzled by the following example where I first convert a date string into a timestamp before converting int back into a date string. However, instead of getting returned "2005:01", gretl returns "2004:12". It seems that the strptime() function does some weird thing, as can be checked here: https://www.epochconverter.com/ But maybe I miss something fundamental here. This is with latest git on ubuntu 19.04. <hansl> string init_obs = "01/2005" scalar tstamp = strptime(init_obs, "%m/%Y") eval strftime(tstamp, "%Y:%m")         # returns '2004:12' </hansl>

On Thu, 25 Apr 2019, Allin Cottrell wrote: > On Thu, 25 Apr 2019, Artur Tarassow wrote: > >> I am a bit puzzled by the following example where I first convert a >> date string into a timestamp before converting int back into a date >> string. However, instead of getting returned "2005:01", gretl returns >> "2004:12". > > I see what you mean. Apparently that's how strptime and strftime work: > if you just specify January 2005 on input strptime gives the epoch > value for midnight of December 2004/January 2005, which seems to > belong to 2004 by convention (of strftime, at least). On further investigation: it seems that (a) the day of the month defaults to zero, and (b) the zeroth day of the month is actually taken to be the last day of the prior month. It's not enough to add a second to the timestamp to push the date from Dec 2004 to Jan 2005: you have to add 24 hours.

I guess we should mention this point in our strptime/strftime documentation. (It's not specific to gretl.)