Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Generate combined with if


From   Nick Cox <njcoxstata@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Generate combined with if
Date   Wed, 10 Aug 2011 16:44:12 +0100

The spacing is 500 somethings. So, translate 30 min and 5 min to
multiples of that and use subscripts for leads and lags.

Or, even better, -tsset- and use time series operators.

Nick

On Wed, Aug 10, 2011 at 4:40 PM, Katia Bobulova
<katia.bobulova@googlemail.com> wrote:

> This is an example of my dataset:
>
> date            time             price_b       S1           S2         id
> 26jan2004       83000   100.095    .              .            1
> 26jan2004       83500   100.125    .              .            2
> 26jan2004       84000   100.12  -0.35          2.08            3
> 26jan2004       84500   100.11                                 4
> 26jan2004       85000   100.125                                5
> 26jan2004       85500   100.17                                 6
> 26jan2004       90000   100.145                                   7
> 26jan2004       90500   100.18                                     8
> 26jan2004       91000   100.16                                 9
> 26jan2004       91500   100.14                                10
> 26jan2004       92000   100.095
> 26jan2004       92500   100.12
> 26jan2004       93000   100.14
> 26jan2004       93500   100.15
> 26jan2004       93500   100.15
> 26jan2004       94500   100.18
> 26jan2004       95000   100.185
> 26jan2004       100000  100.185
>
> i want to construct a regression like this:
>
> reg (price_b(id=10)-price_b(id=3))/price_b(id=3)  S1(id=3)  S2(id=3)
>
> thanks for any help
> Katia
>
> 2011/8/10 Nick Cox <njcoxstata@gmail.com>:
>> I guess subscripting provides at least part of the answer, e.g. [_n -
>> lag1], [_n + lag2].
>>
>> To get better answers, you need to tell us much more about your data.
>>
>> Nick
>>
>> On Wed, Aug 10, 2011 at 4:23 PM, Katia Bobulova
>> <katia.bobulova@googlemail.com> wrote:
>>
>>> I want just to do regressions for different time intervals (identified
>>> by id) and the dependent variable should be the price 30min later
>>> (id=10) and 5minutes before (id=3) the event happen.
>>>
>>> cond doesn't seem to be the solution.
>>
>>> 2011/8/10 Maarten Buis <maartenlbuis@gmail.com>:
>>>> On Wed, Aug 10, 2011 at 4:48 PM, Katia Bobulova wrote:
>>>>> I would like to construct this new price, given by the combination of
>>>>> other prices when there is a particular id.
>>>>>
>>>>> I typed:
>>>>>
>>>>> by date: gen p=(price_b (if id==10) - price_b (if id==3))/ price_b (if id==3)
>>>>
>>>> This does not make sense to me, what should happen when id != 10 or
>>>> when id != 3?
>>>>
>>>> Anyhow, the syntax is not valid as you noticed. Consider looking at
>>>> -help cond()-.

*
*   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/


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