Application of Difference-in-Discontinuity on Household Panel Data
Author: Muhammad Hayat

ABSTRACT

This paper examines the impact evaluation of a program, such programs initiated by the government or non-government authorities or NGOs. The quality of impact evaluation helps to reliable the effectiveness of the programs. In contrast, the impact evaluation also depends on different statistical methodologies to assess variations in outcomes attributed to a proper intervention based on cause and effect analysis. Many econometrics designs or techniques are used as counterfactuals of the impact evaluation. Either these are experimental or quasiexperimental. The average differences between beneficiary and non-beneficiary are the Average Treatment Effect (ATE). Such differences are estimated through multiple techniques widely implemented by the researchers, i.e. Inverse Probability Weighting Regression Adjustment (IPWRA), difference-in-difference (DID), and Regression Discontinuity Design (RDD). Implementing these techniques depends on the nature of the data and the modality of the programs. The main hypothesis and objectives of the study are based on capturing the impact of cash transfer on chronic food security by applying two different models. The study hypothesize that does BISP cash transfer help reduce chronic food insecurity among the benefices, and which technique is more flexible and robust, whether Difference-in-Discontinuity or Inverse Probability Weighting Regression Adjustment (IPWRA). The main objectives of the study are outlined to estimate the impacts of cash transfer on chronic food insecurity among households using Difference-in-Discontinuity and Inverse Probability Weighting Regression Adjustment (IPWRA). The main purpose of the study to compare the estimates of both techniques the Difference-in-Discontinuity and Inverse Probability Weighting Regression Adjustment (IPWRA). The study’s findings explore the positive and statically significant impact of the BISP cash transfer on chronic food insecurity. Treatment Effect Model estimates suggest a strong positive and significant impact of BISP cash transfer on food security outcomes. On the other hand, RD estimation also shows a strong positive and statistically significant impact of BISP on food security outcomes. Hence, the overall results show that cash transfer is helping the poor territory to eradicate chronic food insecurity and move them upward to purchase quality food. Somehow the comparisons of the results of the different models (IPWRA) (RDD) and (Diff-inDis) show that from the rest of the models, the (IPWRA) is more fixable and robust to estimate the impact evaluation of the program.

Meta Data

Keywords : BISP, DID, Difference-in-Discontinuity, Impact Evaluation, IPWRA, RDD, TEM
Supervisor: Ghulam Mustafa

Related Thesis​