Predicting Working Poor and Total Employment in Kenya in-line with SDG norms

Abstract

This research seeks to bridge the gap between the intuitive understanding that jobs are essential for poverty reduction and economic growth. Focusing on Kenya, a Lower-Middle-Income country, this research examines how Total Factor Productivity (TFP) plays a pivotal role in long-term economic growth. This research lists all the key determinants contributing to the enhancement of TFP in-line with SDG 1, 2 & 8 and investigates the complex interrelationships between Contribution to GDP and the Growth of GDP to predict Working Poor and Total Employment for each industry. The dataset for analysis was accumulated from the Statistical reports from 2011 to 2023, published by the Kenya National Bureau of Statistics (KNBS). This empirical research employs Deep Learning (DL) and Machine Learning (ML) algorithms to predict the number of Working Poor and Total Employment, based on economic growth metrics. After comparing various ML and DL algorithms, the research concludes that the Decision Tree is the most suitable for predicting the Working Poor with 94.76% accuracy, while the Random Forest is most effective for predicting Total Employment with 94.57%. These predictions are intended to assist policymakers in making informed decisions to reduce the Working Poor, thereby increasing TFP and, in turn, bolstering the Kenyan economy.