This repository includes the scripts required to replicate the main findings of the article "Inequality of the crowding-out effect of tobacco expenditure in Colombia" published in PLOSONE. We include a minimum dataset for the replicability of tables and figures, but the base datasets can be obtained from the ANDA repository: https://microdatos.dane.gov.co/catalog/MICRODATOS/about. If you need something more detailed on the construction of the data, please contact us.
- Folder scriptsECV take base ECV files and generates standardized versions of them for this paper. For instance, health expenditures and income. Needs original datasets from DANE.
- ECVREP_01_compile.do and ECVREP_02_matchData.do generate the file. Needs outputs from the previous points. These files produce "ECVrepeat_preMatch.dta". This file can be downloaded here:
- ECVREP_02b_matchData.r produces de genetic matching and generates "ECV_pesosAfterMatchingSmokers.dta"
- ECVREP_03_SUREanalysis.do compiles the two datasets described above into "$mainF\ECVrepeat_a8.dta". With this file is possible to run the SURE analyses.