[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

From |
"Nick Cox" <n.j.cox@durham.ac.uk> |

To |
<statalist@hsphsun2.harvard.edu> |

Subject |
st: RE: time series problem |

Date |
Wed, 17 Jul 2002 17:53:08 +0100 |

Michael Carney > I am trying to analyze data from the dow jones cash market > and the dow > futures market. i am having the following problem: > > may trading program gives me data that runs in thusly, for example: > > TIME PRICE > 10:30 8250 > 10:30 8253 > 10:30 8248 > 10:31 8249 > 10:31 8250 > > The time does not include colons and does not pose a > problem. What is > vexing, however, is that Stata cannot convert the data to a > time series set > as my trading program does not differentiate between > different values for > 10:30 AM, for example; exporting the data to a text file > and pasting in > Stata's data editor I have many different values for what > appears to be > 10:30, for example, but is actually 10:30 and so many seconds. > Unfortunately, I cannot get copyable data second by second > from my trading > program (TradeStation). Is there any way I can code all of > the 10:30 > values, for example, so that i can get the first datum for > 10:30, the > second, third, etc. coded in order to do an adequate time > series analysis? > I want to analyze the market tick by tick and plot a day's worth of > statistics and run regressions with other markets. ANY help > would be > greatly appreciated. > You can generate two variables which are just gen time = _n and something like, assuming that TIME is string, although I am not clear where the colons come in, gen str1 TIME2 = "" bysort TIME (time) : replace TIME2 = TIME + ":" + string(20 * (_n-1)) You could do computations using -time- and present them using TIME2. Something like this may be essential anyway as Stata does not support time of day as such. Nick n.j.cox@durham.ac.uk * * For searches and help try: * http://www.stata.com/support/faqs/res/findit.html * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/

**Follow-Ups**:**st: Checking status during long calculations***From:*Felice Martinello <fmartine@taro.bus.brocku.ca>

**References**:**st: time series problem***From:*Michael Carney <carneymh@earthlink.net>

- Prev by Date:
**Re: st: comparing anova and regression results** - Next by Date:
**Re: st: Population size in survey data** - Previous by thread:
**st: time series problem** - Next by thread:
**st: Checking status during long calculations** - Index(es):

© Copyright 1996–2014 StataCorp LP | Terms of use | Privacy | Contact us | What's new | Site index |