Re: st: High-frequency time-series (stata 10)

["Marc N. Sch�ssler" <marc.schuessler@medma.uni-heidelberg.de>]

Hi Bea,

i had a similar problem, assigning activity-levels to 15 sec.  
Intervalls.
You need to have 5 min- intervalls in long format and within these 5  
min. intervalls the different bid and answers in wide format.

My problem had been similar, an activity - level was measured and  
could change within a 15 second Intervall.
So i devided my  timevar in 15 seconds pices starting here with  
7:00am as observations begin.
Over a loop then assign a new variable the clocktimevalue in  
milliseconds  ( i called it "time15").
then i looped over every second from this 15 second Intervall and  
assigned a counting device for the reshape to keep the ordering  
within the 15 second Intervalls

In the end i droped unneeded variabls and then reshaped by (id  
time15) to keep the ID-TIME order and j = alevel15 to get within the  
15 second intervall the different activitylevels into the wide format.

hope this helps, however i noticed that there is some trouble with  
the charactercoding switching from MAC to PC. So you might see the  
special characters like brakets for the loop not properly.

greets
marc

**********************************************************************
cap drop alevel15 time15
gen time15 = .
gen alevel15=.
local help = 25200000						///this means 7:00am in milliseconds
forvalues x=  25200000(15000)48600000	 {
	disp "`x'"
	bysort ID: replace time15 = `help' if time >= `x' & time < (`x' +  
15000)
	local help = `help' + 15000
		forvalues y = 0/14 {
			replace alevel15 = `y' if time==`x' + `y'*1000
		}
	}

keep ale* ID  time15
	reshape wide alevel ,i(ID time15) j(alevel15)

**********************************************************************


Am 24.09.2008 um 13:33 schrieb Beatrice Crozza:


> Dear All,
>
> I know that with Stata 10 all time-series analysis commands now
> support data with frequencies as high as 1 millisecond.
>
> I would like to treat my dataset as a time-series,I have
> high-frequency data(5-minute intervals) and a variable (timedate) that
> is composed of the time and the date for each bid and ask. My problem
> is that I have duplicate data, because for each timedate there could
> be more than one bid and more than one ask, so Stata gave me this
> error message:
>
> . tsset timedate, delta(5 minute)
> repeated time values in sample
> r(451);
>
> I need to treat my date as a time-series to do an AR analysis for bids
> and asks through time.
> I am thinking to assign a number ( in sequential order) to each
> timedate and than to use the command tsset with this new variable.
>
> Do you think that this procedure is right to achieve my goal?
>
> Do you have any idea on how to assign a sequential number to each bid
> and ask, without using Excel?
>
> Thank you very much in advance.
>
> Best,
> Bea
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/