Skip to content
No description, website, or topics provided.
R C++ Rebol
Branch: master
Clone or download
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
.gitignore README updated Jan 8, 2016
AM2011Table1cols2356.R Arcidiacono and Miller (2011) upload Jan 8, 2016
BayesianDDCEstimateDataRustvEmax.R
EstimateBayesianHierarchicalDDC.R Removed spaces from filenames Feb 14, 2016
EstimateBayesianHierarchicalDDCcpp.R Removed spaces from filenames Feb 14, 2016
Import Data and Estimate.R README updated Jan 8, 2016
README.txt Removed spaces from filenames Feb 14, 2016
Rust Data Generating Process.R
RustDGPwithHierarchicalRE.R Removed spaces from filenames Feb 14, 2016
bddcMCMCloop.cpp
bus_df_lin.csv
clogit.R README updated Jan 8, 2016
fvdataBOTH.cpp
genbus4.cpp Arcidiacono and Miller (2011) upload Jan 8, 2016
intcond.R Arcidiacono and Miller (2011) upload Jan 8, 2016
intcondP.R Arcidiacono and Miller (2011) upload Jan 8, 2016
likebusML4.R Arcidiacono and Miller (2011) upload Jan 8, 2016
npl_sing.R README updated Jan 8, 2016
single-agent-dynamic-choice.Rproj README updated Jan 8, 2016
wlogit.R Arcidiacono and Miller (2011) upload Jan 8, 2016
wlogitd.R Arcidiacono and Miller (2011) upload Jan 8, 2016
xgrid.R

README.txt

This folder contains three groups of self-contained files:

First Group: Nested Fixed Point and Nested Pseudo Likelihood------------------
1) "Rust Data Generating Process.R” illustrates a simple DGP for the Rust bus dataset.
   It generates the data and estimates the parameters using a nested fixed point algorithm
   The output file “bus_df_in.csv” is used in the Nested Pseudo Likelihood algorithm

2) "Import Data and Estimate" implements Aguirregabiria and Mira’s (2002) NPL algorithm
   It imports “bus_df_in.csv” generated above and calls "npl_sing.R" and "clogit.R"

Second Group: Arcidiacono and Miller (2011)------------------------------------
1) “AM2011Table1cols2356.R” recreates columns 2,3,5 and 6 from Arcidiacono and Miller (2011)
   Data is simulated within the program
   It calls the following support functions:
   -xgrid.R,wlogitd.R,wlogit.R,likebusML4.R,genbus4.cpp,fvdataBOTH.cpp,intcond.R,intcondP.R

Third Group: Bayesian DDC (2009)------------------------------------------------
*) "BayesianDDCEstimateDataRustvEmax.R" uses the output from "Rust Data Generating Process.R” to
replicate the Imai, Jain, and Ching (2009) method, but does not allow for random effects. Use this
file to get a base understanding of the model before proceeding to the hierarchical version.

1) "RustDGPwithHierarchicalRE.R" generates Rust data with a hierarchical mixing

2) "EstimateBayesianHierarchicalDDC.R" follows a method similar to Imai, Jain, and Ching (2009)
for estimation. See the code for options on variations of the estimation procedure.

3) "EstimateBayesianHierarchicalDDC cpp.R" is similar to #2 but uses c++ for speed improvements.
The main c++ program is in "bddcMCMCloop.cpp"

IMPORTANT: this is a stylized example meant to highlight the mechanics of the process. Care must be
taken when selecting the priors and scaling parameters. Here convergence is achieved, but only because
of how I set up the problem.
You can’t perform that action at this time.