Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: Filling Missing Times/Dates and Corresponding Variable Entries
From
Nick Cox <[email protected]>
To
[email protected]
Subject
Re: st: Filling Missing Times/Dates and Corresponding Variable Entries
Date
Thu, 26 May 2011 10:03:30 +0100
This is surely a standard problem with this kind of data. I don't
think it's like an interpolation problem in which there is a gap
because someone forgot to take a measurement or the record is
illegible or whatever, as is common in meteorology, for example. One
answer is that the clock stops when no-one is trading, so you
shouldn't interpolate: you just work in terms of business time. See
SJ-7-1 dm0028 . . . . . . . . . . . . Stata tip 40: Taking care of business
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. F. Baum
Q1/07 SJ 7(1):137--139 (no commands)
tip for dealing with time-series date schemes and how to
handle gaps
for a discussion. If you do interpolate, it messes up your dependence
structure, not to mention the spectrum. Either way, examiners and/or
reviewers would expect you to be familiar with what is standard in
your field, rather than invent something ad hoc.
Nick
On Thu, May 26, 2011 at 9:20 AM, Muhammad Anees <[email protected]> wrote:
> I have the following examplified stock market data on three different
> companies in different csv files for the same times or dates in each
> sheet:
>
> time share price market capitalization
> volume
>
> 01JAN2005 20 2000
> 10022
> 02JAN2005 21 2001
> 10023
> 03JAN2005 22 2002
> 10024
> 04JAN2005 23 2003
> 10025
> 05JAN2005 24 2004
> 10026
> .
> .
> 08JAN2005 26 2007
> 10029
> 09JAN2005 27 2008
> 10030
> 10JAN2005 28 2009
> 10031
> 11JAN2005 29 2010
> 10032
> 12JAN2005 30 2011
> 10033
> .
> .
> and so on where the missing date is either a weekend or public holidy.
>
> Now my problem is how to fill the time gaps and the corresponding
> values for given variables using either values on the prior dates of
> each variable or average of the last and next working dates. I know
> stata offers ways to fill missing values but how can I do it in this
> case.
>
> Thank you very much in advance for guiding me through the problem.
*
* 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/