Click here for search results

Poverty Analysis Toolkit

Poverty and Inequality Research

This page Includes Stata programs (all of which work under STATA 8/9) for poverty dynamic analysis. Several other programs of general interest for economist are also available.

Poverty Analysis 
(Stata programs)
Applied Economic Research 
(Stata programs)

Researchers
Michael Lokshin  | Martin Ravallion  | Zurab Sajaia (email: ZSajaia@worldbank.org

 

 

 

 

 

  •  
  •  

 

 

 


Poverty Analysis Toolkit

Michael Lokshin and Martin Ravallion are developing a set of Stata ado programs to make accessible various methodologies for dynamic poverty analysis. 

As of today, the Poverty Analysis Toolkit includes four programs for dynamic poverty analysis. The first two programs GIDECOMPOSITION and SEDECOMPOSITION perform growth-inequality and sectoral decompositions of poverty between two periods of time. Programs GICURVE and POV_ROBUST produce graphs. The first program produces the growth incidence curve and the second program produces various graphs for stochastic dominance analysis. The programs are accompanied by the help files with a detailed explanation of the programs’ syntax and applications. The sample program “Dynamic_poverty_example.do” illustrates the use of the program on a fictitious dataset.



All programs are designed to work under Stata 8/9. In order to use the programs copy ADO and HLP files into Stata's Ado directory. This directory is usually C:/ado. Alternatively, start Stata and type "adopath" to see the directory where Stata keeps its ado files and copy files in that directory.

Alternatively, you can download the programs by linking to the website within Stata by typing:

.net from http://siteresources.worldbank.org/INTPOVRES/Resources/

Note: Stata is case-sensitive so you must type the URL address as it is given here

tool-backtotop.gif


Growth-inequality decomposition (latest version 05/15/07):

Within Stata type:

 .net install gidecomposition, replace from(http://siteresources.worldbank.org/INTPOVRES/Resources)

tool-backtotop.gif


Sectoral decomposition of inequality (latest version 05/15/07):

Within Stata type:

 .net install sedecomposition, replace from(http://siteresources.worldbank.org/INTPOVRES/Resources)

tool-backtotop.gif


Growth Incidence Curve: Updated 05.23.2007

Within Stata type:

 .net install gicurve, replace from (http://siteresources.worldbank.org/INTPOVRES/Resources)

tool-backtotop.gif


Stochastic Dominance Analysis

Within Stata type:

 .net install pov_robust, replace from (http://siteresources.worldbank.org/INTPOVRES/Resources)

tool-backtotop.gif


Dynamic Poverty Example

egen extention functions for inequality and poverty measures:

Within Stata type:

 .net install egen_inequal, replace from (http://siteresources.worldbank.org/INTPOVRES/Resources)

tool-backtotop.gif


Fast algorithm for calculation of the Gini coefficient and it's jackknife standard errors

Within Stata type:

 .net install fastgini, replace from (http://siteresources.worldbank.org/INTPOVRES/Resources)

tool-backtotop.gif


Stata programs for Applied Economic Research

In addition to the poverty analysis toolkit, we developed several programs of general interest for economists.

MOVESTAY program implements a Full Information Maximum Likelihood (FIML) algorithm to estimate the endogenous switching regression model.

Within Stata type:

 .net install movestay, all replace from(http://siteresources.worldbank.org/INTPOVRES/Resources)

Note: the program movestay has been updated in June 2006, please reload the new files.

tool-backtotop.gif


OHECKMAN - Selection model based on ordered probit

Within Stata type:

 .net install oheckman, replace from (http://siteresources.worldbank.org/INTPOVRES/Resources)

tool-backtotop.gif


BIOPROBIT  performs maximum-likelihood estimation of two-equation ordered probit models with endogeneity. For more detailed description of the model please see bioprobit.pdf.

To install the program, within Stata type:

findit bioprobit

and follow the instructions there.

tool-backtotop.gif


PLREG semi-parametric difference-based estimation of partial linear regression models.

Partial Linear Regression Model:

Within Stata type:

 .net install plreg, all replace from(http://siteresources.worldbank.org/INTPOVRES/Resources)

Note: the program plreg has been updated in August 2006, please reload the new files

tool-backtotop.gif


BESTREG is a procedure of selecting a subset of variables providing the model's best fit. It implements "Leaps and Bounds" algorithm for optimal subset search.

Within Stata type:

 .net install bestreg, replace from (http://siteresources.worldbank.org/INTPOVRES/Resources)

tool-backtotop.gif


PPREG does efficient estimation of regressions based on pseudo-panel data.

Within Stata type:

 .net install ppreg, replace from (http://siteresources.worldbank.org/INTPOVRES/Resources)

tool-backtotop.gif


LOOKFOR_ALL is an extention of a Stata program lookfor. lookfor_all searches through all Stata data files in a current directory and subdirectories
for the variable(s) with a specified pattern. The pattern could be a part/whole variable name and/or a word(part of the word) in the variable label.

To install the program, within Stata type:

findit lookfor_all

and follow the instructions there.

tool-backtotop.gif


XML_TAB is a program that saves the outputs from Stata estimation procedures in Microsoft Excel.

To install xml_tab, within Stata type:

 findit xml_tab

and follow the instructions there.

tool-backtotop.gif


Programs (in Matlab) for the paper by Korinek, A., Mistiaen J., and M. Ravallion (2006) "An Econometric Method of Correcting for unit Nonresponse Bias in Surveys" forthcoming in Journal of Econometrics:

tool-backtotop.gif




Permanent URL for this page: http://go.worldbank.org/YF9PVNXJY0