Pakistan Poverty Map 2019–2020

This paper summarizes the approach used to estimate monetary poverty for Pakistan in 2019–20 at the district level. The small area estimation method is used to impute welfare from the Household Income and Expenditure Survey 2018–19 into the Pakistan Social and Living Standard Measurement Survey 2019–20. This application differs from the standard small area estimation implementations that use a survey-to-census method because the two surveys are household surveys. Using surveys instead of a census as the target data offers additional information for modeling poverty, but it also comes at the expense of noise due to sampling. Monetary poverty rates are estimated for 126 districts in Punjab, Sindh, Khyber Pakhtunkhwa, and Balochistan, including, for the first time, the districts of the former Federal Administered Tribal Areas and the former Frontier Regions. Using the Census Empirical Best method, the analysis obtains poverty estimates with higher precision and accuracy than those of the methodology previously implemented in Pakistan.

本文总结了2019—2020年巴基斯坦区级货币贫困的估算方案。本研究采用小区域估计（small area estimation）方法，将2018—2019年住户收入与支出调查（Household Income and Expenditure Survey）中的福利水平数据，推算至2019—2020年巴基斯坦社会与生活标准衡量调查（Pakistan Social and Living Standard Measurement Survey）中。该方案不同于标准小区域估计的常规实现流程——后者多采用调查-普查映射方法，而本研究的两项数据源均为住户调查。以住户调查而非普查作为目标数据集，可为贫困建模提供额外信息，但同时也会因抽样引入噪声误差。本次估算覆盖旁遮普省、信德省、开伯尔-普赫图赫瓦省及俾路支省共计126个区县，首次纳入原联邦直辖部落地区与原边境地区的所有下辖区县。本研究通过普查经验最优法（Census Empirical Best method）开展分析，所得贫困估算结果的精确度与准确度均优于巴基斯坦此前应用的相关方法。