LetRegNum(8,11);

Save  the file and the run the installation macro. Of course, this will break once version 12 comes out, which is normally about a month after I decide to splurge on an upgrade.  Wouldn’t it be nice if Statacorp gave you a warning that a new version was on the horizon?  I know, I should be using R.  It also plays nice with (R)Winedt.

Stata Mode lacks one convenient feature – the ability to run only selected lines of your Stata code.  This can be achieved by following these instructions.

• Download gnuplot and unzip in some folder, e.g., C:\gnuplot.  Subsequently add C:\gnuplot\bin\ as a path to your environment variables.
• If you have downloaded the windows version of gnuplot then change the name of the executable from ‘wgnuplot.exe’ to ‘gnuplot.exe’.
• gnuplot may be prevented from writing in the current directory when it is called by tikz.  Use the –enable-write18 switch when compiling your document, e.g., pdflatex –enable-write18 foo.tex.  If you use WinEdt, the switch can be added under Options-Executions Modes.   Note that the first ‘-’ in –enable-write18 should be two ‘-’.
• gnuplot doesn’t seem to like spaces in filenames.

Journal article:
Kydland, F. E., Prescott., E. C. (1977) Rules rather than discretion: The inconsistency of optimal plans. Journal of Political Economy 85(3): 473–49

Complete book:
Kuran, T. (1995) Private Truths, Public Lies: The Social Consequences of Preference Falsification. Harvard University Press, Cambridge

Contribution in a book:
Schelling, T. (1968) The life you save may be your own. In: Chase, S. B. (ed.) Problems in Public Expenditure Analysis. The Brookings Institution, Washington, D.C.

The bibliography style: ecogov.bst

data [stata]

codebook [pdf]

I didn’t collect the data so please head over to Assisting Politics if you have comments, etc.

The MIPT Terrorist Knowledge Base (www.tkb.org) provides data on international terrorist incidents since 1968 and domestic terrorist incidents since 1998. The database is updated monthly. Each terrorist incident can be looked up individually online but the whole database is not distributed publicly. In a recent paper I examined the effects of terrorism on coalition formation (Indridason, Indridi H. “Does Terrorism Influence Domestic Politics? Coalition Formation and Terrorist Incidents.” Forthcoming 2008. Journal of Peace Research 45(2)) and to obtain the data I wrote a perl-script that flips through each page and retrieves that data. Dave Armstrong deserves credit for giving me the idea to write the script. Even if you are not interested in terrorist data, the script should be fairly easy to adopt to download any data on the web – it only took me a couple of days to write the script without having ever touched perl before.

terror perl script

Coalition data – Iceland 1945-2003 [.xls]

If you have data that you would like to share with others I will be happy to post your data here or provide a link to where your data is stored. If you want to share your data on this site you will need to send me:

• the data
• a codebook with details on sources, variables, coding, etc.
• information about how you would like potential users of your data to cite you
• some keywords, e.g., geographic region, country, thematic (elections, legislatures, etc.)

Note that there are other places that you may want to deposit your data, e.g., at the IQSS Dataverse Network or ICPSR, and are perhaps more appropriate for larger datasets. The aim of this site is not too compete with these institutions but rather provide access to `basic’ data the researchers often have a hard time gathering. That is, my motivation for running this site stems from the fact that countries such as my homeland, Iceland, are often excluded from comparative studies because of lack of data. Often the data has been collected or exists but hasn’t been made accessible. The hope is that this site will ease the search for such data and encourage students and scholar to share their data here instead of keeping it buried in boxes in the corner of their offices.