I have a panel dataset with dates and times in the seconds. It comes from a text file in the format HH:MM:SS, converted with the clock function.
The time from the text files is local time, not GMT, and the time shifts with the beginning and end of daylight savings time. I am not doing any regressions or the like with the data, but I need to take first differences of some variables. Currently, I am not using tsset and instead using [_n] and [_n-1] making use of if statements regarding the time of year (DST or non-DST) and the changeover between the two (while also including a statement to make sure I don't difference across panels). Is there anything built-in or available for Stata that accounts for DST status that would allow me to use tsset functions such as d.?
Thanks
The time from the text files is local time, not GMT, and the time shifts with the beginning and end of daylight savings time. I am not doing any regressions or the like with the data, but I need to take first differences of some variables. Currently, I am not using tsset and instead using [_n] and [_n-1] making use of if statements regarding the time of year (DST or non-DST) and the changeover between the two (while also including a statement to make sure I don't difference across panels). Is there anything built-in or available for Stata that accounts for DST status that would allow me to use tsset functions such as d.?
Thanks
