Bookmark and Share

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]

st: Time series analysis with data with gaps


From   "Ikuho Kochi" <[email protected]>
To   statalist <[email protected]>
Subject   st: Time series analysis with data with gaps
Date   Thu, 20 Feb 2014 21:59:07 +0000

Dear fellow STATA users.

I am trying to run a time series model with the data with gaps.
 
The data contains daily observations between October 6 to November 16 from different years.
Time variable is created based on Stata Internal Form (SIF) of the date of observation.

For example, my first observation comes from October 6 in 2005, and thus has t=16715. 
The data contains consecutive observations until t=16756 (Nov 16, 2005).  
Then the following observation is from Oct 6, 2006 with t=17080 and contains consective observations until t=17121 and so on.

When there is such huge gaps in data, what is the best way to handle autocorrelation? 

The model contains dummy variable for each year to control the fact that data comes from 
different years and I also include "t" variable.

When I "tsset" data using "t" variable, STATA recognized that there are gaps in data.

Now, I am wondering if I can use "newey" command to obtain Newey-West standard errors to correct potential autocorrelation
 in my model where my data contains such big gaps.  In "newey" command, does the program assumes all data are consecutive,
although the t variable indicates otherwise?, or it takes into account the gaps and calculate valid standard error?


I welcome any suggestions or advice.

Thank you very much.

Ikuho Kochi
*
*   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/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index