First of all, thanks to Naheed Mirza for the kind reply.
However, I had already used '. tsset time' in my commands (see my applied
commands below this e-mail), but didnot work out (may be due to gappy date
- is there any way to handle such quarterly date?) and unable to find
proper help for command -tdates- (i'm a newbie - i tried help for tdates in
stata 7 help menu, but using '. format mydate %tq' don't seem to work)!
Any response / help / comment / suggestion / idea / web-link / replies will
be greatly appreciated.
Mohammad Ehsanul Karim <email@example.com>
Institute of Statistical Research and Training
University of Dhaka, Dhaka- 1000, Bangladesh
>>> Your data are quarterly. Best to -tsset- them as such. For
example, see help on -tsset- and -tdates-.
. infix str10 YEAR 1-10 GDP 11-20 PDI 21-30 PCE 31-40 PROFITS 41-50
DIVIDENDS 51-59 using c:\gujarati.txt
. generate times = date(YEAR,"dmy")
. format times %dD.N.CY
. label data "Macroeconomic Data, United States, 1970.1 to 1991.4"
. note: Data taken from Gujarati (2003) 'Basic Econometrics' 4 th ed. Page 794
. save c:\gujarati.dta
/* This is saved in http://www.angelfire.com/ab5/get5/gujarati.dta */
. use c:\gujarati.dta
. gen day = day(times)
. gen month = month( times)
. gen year = year( times)
. gen mydate = mdy(month,day,year)
. format mydate %tq
. tsset mydate
time variable: mydate, 2873q2 to 4836q1, but with gaps
. corrgram GDP, lags(25)
(note: time series has 86 gaps)