Changes

Jump to navigation Jump to search
}}
=Current Work=
 
==How to do a run==
 
===Before you begin===
 
All of the following assumes that you're working on Ed's infrastructure. If you're not, you'll need to set up a dbase server, and load up the '''lbo''' dbase, either by restoring it from a backup (see below) or by running Buyouts.sql and having the appropriate source files available (in the zip "sourcefiles.zip").
 
To restore the backup to a postgres server
createdb lbo
pg_restore -Fc -d lbo lbo_Fc.dump
 
To build it from source files:
unzip sourcefiles.zip
createdb lbo
psql lbo
\i Buyouts.sql
 
===A run===
 
Buyouts.sql produces Masterv2-2.txt. If you have this in hand, you'll need the following to do a run:
*Datasetup.do
*LBOmatchscript.jl and LBOmatching.jl
*The '''lbo''' database loaded up on a postgres dbase server
*PatentMeasures.sql
*Some static STATA files: CPI.dta, OldData.dta.
 
The sequence is then:
1 Run Datasetup.do
do Datasetup
2 Do the matching
julia -i LBOmatchscript.jl
println("Number of LBOs with complete data available for matching ", size(LBOs,1))
diagdf
exit()
3 Create the patent measures
psql lbo
\i PatentMeasures.sql
4 Run the analysis
do Analysis.do
5 Wonder whether this was all worth it.
6 Edit the excel files (some are xml files that have to be opened from inside of Excel), optionally moving columns up to create alignment and using regexes to get rid of unwanted rows. Create and apply a chart template to simplify the standardized production of figures. Use the excel2latex plugin to export tables to LaTeX.
==Review of Status Quo==

Navigation menu