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: business days
From
Nick Cox <[email protected]>
To
"[email protected]" <[email protected]>
Subject
Re: st: business days
Date
Wed, 4 Dec 2013 19:25:33 +0000
1. Stata has business calendars. Start with -help bcal-.
2. I don't think time series data raise any -merge- issues that aren't
generic. So, -merge- and -drop- is a reasonable answer to this
question. If you want a more detailed answer, you need to ask a more
specific question showing what you tried and what is problematic about
it.
3.
a) Already documented at
http://www.stata.com/support/faqs/data-management/replacing-missing-values/
b) Already documented at help for -ipolate-.
But the whole point of a business calendar is that gaps are to be
treated as not gaps, removing the need for interpolation (I don't use
the word "imputation" here).
Advice: You ask for a patient and thorough answer. We ask that you
heed the advice at
http://www.stata.com/support/faqs/resources/statalist-faq/#before
Nick
[email protected]
On 4 December 2013 19:11, Mats Ljungren <[email protected]> wrote:
> Hi,
> I am using finance time series such as equity prices from different countries with different holidays vs business days.
>
> I need help with three issues,
> 1. How can I make sure Stata to understand that my I have a daily series including only business days?
> 2. How to merge series and delete generated missing observation from the new data set?
> 3. How may I do imputation for missing observations, a) let the observation before the missing observation to be repeated and replace the missing observation, b) replace the missing observation with the average value of the observations before and after. In the latter case maybe there are 2-3 holidays that must be filled in with the average of the observation before the missing series and the observation after the missing series.
>
> Hopefully the problems are well known for many of you.
>
> I will thank you in advance for helping me thoroughly and patiently.
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/