Poverty gaps and headcount/incidence

[Chei Bukari]

Dear All,
I am currently examining the relative and joint effects of complementary financial services on household poverty in Ghana.
I hypothesized that, households with multiple/complementary financial service (i.e., credit*insurance, credit*saving, credit*insurance*savings, etc) would have higher expenditures than their counterparts who do not consume these multiple product. I am using a survey data (GLSS 6) which has poverty line in it. The data also contains incomes and various types of expenditure.
I am currently not sure of the stata command that would help me calculate poverty gaps and poverty head-counts across these two groups (i.e. those with multiple services against those without)
can any one sort me out?

Again, does Stata calculate this using the FGT approach or which specific model and or formula does stata use in computation of poverty gaps and head-counts?

[Stephen Jenkins]

A little bit of searching would have revealed some Stata programs for calculating poverty indices including those in the Foster-Greer-Thorbecke (FGT) class. Inter alia, I can recommend povdeco on SSC. If you wish to also calculate standard errors for poverty index estimates while also taking account of survey design, then you could simply apply Stata's svy commands to a suitably-defined individual-level 'deprivation' indicator (a function of the poverty gap). For worked examples, see the presentation included in the following SSC package: Jenkins, S.P. 2006. Estimation and interpretation of measures of inequality, poverty, and social welfare using Stata. Presentation at North American Stata Users' Group Meetings 2006, Boston MA. http://econpapers.repec.org/paper/bocasug06/16.htm.

[achintya widhi]

Dear all, i want to calculate p0, p1, and p2. My data is cross section year 2014, households and individuals survey data. my variabels are percapita expenditure, province, age, agesquare, education, gender, landownership. how can i calculate P0, P1, and P2? because those indexes will be used as outcomes variabel to running propensity score matching.