Does the control variable must always be a non-user of irrigation for PSM to work?
https://www.researchgate.net/publication/278670630_Can_Zakat_System_Alleviate_Rural_Poverty_in_Bangladesh_A_Propensity_Score_Matching_Approach
Article Can Zakat System Alleviate Rural Poverty in Bangladesh? A Pr...
Hi Mia. No us the short answer. First just model farm income with irrigation as a right hand side variable. Then estimate a different set of models which allow for different types of farmers to choose different irrigation. Here you estimate step 1 which is the probability that a farmer with different characteristics will choose irrigation method x,y,z. You save these (latent) probabilities for each irrigation type. Then you model farm income with the (latent) probability of using each type of irrigation on the right hand side. Hope this helps? Message me if you need some code. Cheers, Marc
you can also analyze a difference of income of farmers before using three different irrigation system and after using three different irrigation system and find mean score...
- Badges
- Science topic
- Similar topics
- Phytochemistry
- Extracts