Is Random Forest a Superior Methodology for Predicting Poverty? : An Empirical Assessment

Random forest is in many fields of research a common method for data driven predictions. Within economics and prediction of poverty, random forest is rarely used. Comparing out-of-sample predictions in surveys for same year in six countries shows t...

Full description

Bibliographic Details
Main Authors:	Sohnesen, Thomas Pave, Stender, Niels
Format:	Working Paper
Language:	English en_US
Published:	World Bank, Washington, DC 2016
Subjects:	PREDICTIONS POOR HOUSEHOLD CONSUMPTION EXPENDITURES HOUSEHOLD SIZE HOUSEHOLD SURVEY AGRICULTURAL GROWTH CONSUMPTION POVERTY REDUCTION IMPACT ON POVERTY POVERTY RATES ERRORS FARMER POVERTY RATE FOOD CONSUMPTION INCOME LINEAR REGRESSION POVERTY ESTIMATES ALGORITHMS HOUSEHOLD SURVEYS PROGRAMS CONSUMPTION DATA HOUSING AGRICULTURAL PRACTICES IMPACTS NATIONAL POVERTY SAMPLES RURAL VARIABLES MEASUREMENT COUNTING HOUSEHOLD BUDGET CONSUMPTION AGGREGATE QUALITY SURVEYS SOCIAL ASSISTANCE MEASURES INSTRUMENTS TARGETING RANDOM SAMPLES CONSUMPTION EXPENDITURE RURAL AREAS CROSS‐SECTION DATA WELFARE MEASURES WELFARE INDICATORS PANEL DATA SETS REGIONS STATISTICS EVALUATION SIGNIFICANCE LEVEL POOR HOUSEHOLDS SAMPLING POVERTY HOUSEHOLD HEAD HOUSEHOLD CONSUMPTION ECONOMETRICS STANDARD ERRORS POVERTY STATUS POOR PREDICTION POVERTY ASSESSMENT LEARNING INDICATORS RESEARCH CONSUMPTION POVERTY OUTCOMES SOCIAL INDICATORS MISSING OBSERVATIONS INEQUALITY
Online Access:	http://documents.worldbank.org/curated/en/2016/03/26089791/random-forest-superior-methodology-predicting-poverty-empirical-assessment http://hdl.handle.net/10986/24154

Description
Summary:	Random forest is in many fields of research a common method for data driven predictions. Within economics and prediction of poverty, random forest is rarely used. Comparing out-of-sample predictions in surveys for same year in six countries shows that random forest is often more accurate than current common practice (multiple imputations with variables selected by stepwise and Lasso), suggesting that this method could contribute to better poverty predictions. However, none of the methods consistently provides accurate predictions of poverty over time, highlighting that technical model fitting by any method within a single year is not always, by itself, sufficient for accurate predictions of poverty over time.

Is Random Forest a Superior Methodology for Predicting Poverty? : An Empirical Assessment

Similar Items