Household and contextual indicators of poverty in the Gauteng province of South Africa: A multilevel analysis

Abstract

This study aims to analyse the socio-economic factors contributing to poverty reduction in South Africa using time series data from 2006 to 2019. The stationarity of the variables will be assessed by applying the Augmented Dickey Fuller (ADF) test. The Autoregressive Distributed Lag (ARDL) analytical technique will be adopted to analyse the cointegration amongst variables pertaining to different orders of cointegration amongst lower bound [I(0)] and upper bound [I(1)]. The study will analyse the long-term and short-term effects of the socio-economic factors contributing to poverty reduction in South Africa. If the calculated F-statistic is greater than the upper bound [I(1)], the Error Correction Model will be adopted to assess the short-run effects. Diagnostic tests will be performed to test the robustness of the model. The tests will performed will include: (1) the Breusch-Godfrey test for serial correlation; (2) the Jarque-Bera test for normality; (3) the Breusch-Pagan-Godfrey technique to test for heteroscedasticity; and (4) the cumulative sum chart to detect deviation from the average associated with a subgroup.