Time Series Princeton University（时间序列普林斯顿大学）.pdf

Time Series (ver. 1.5) Oscar Torres-Reyna Data Consultant otorres@ /training/ PU/DSS/OTR Date variable For ‘date1’ type: gen datevar = date(date1,DMY, 2099) format datevar %td /*For daily data*/ For ‘date2’ type: gen datevar = date(date2,MDY, 2099) format datevar %td /*For daily data*/ For ‘date3’ type: tostring date3, gen(date3a) gen datevar=date(date3a,YMD) format datevar %td /*For daily data*/ For ‘date4’ see next page Date in integers and unbalanced use /~otorres/Stata/date.dta, clear drop date1 date2 rename date3 date4 *Year, month, day tostring date4, gen(date4a) gen len = length(date4a) gen year = substr(date4a,1,4) *When len=6, month is in 5th position and day in 6th gen month = substr(date4a,5,1) if len == 6 gen day = substr (date4a,6,1) if len == 6 *When len=7 is hard to distinguish month/day, we skip *When len=8, month is in 5th/6th positon and day in 7th/8th replace month = substr (date4a,5,2) if len == 8 replace day = substr(date4a,7,2) if len == 8 destring month day year, replace *Creating datevar gen datevar = mdy (month,day,year) format datevar %td *Filling in the missing dates replace datevar = datevar [_n-1] + 1 if datevar == . Date variable (cont.) If the original date variable is string (i.e. color red): gen week= weekly(stringvar,wy) gen month= monthly(stringvar,my) NOTE: Remember to format the date variable accordingly. After creating it type: gen quarter= quarterly(stringvar,qy) format datevar %t? /*Change ‘datevar’ with your date variable*/ gen half = halfyearly(stringvar,hy) Change “?” with t