Building Prediction Models for Maternal Mortality Rate in Bangladesh

Abstract

Maternal Mortality Rate (MMR) is one of the serious concerns in developing countries particularly in Bangladesh on which the world has an eye on the factors responsible for the causes. The safety of mother before and after giving birth becomes increasingly important in recent times. Several studies have already been done on investigating the reasons for the maternal mortality and various techniques have been conducted to predict the causes of the issues. In this paper, we aim to identify additional risk factors and observe factors identified in various studies related to maternal mortality rate in Bangladesh, using a different data repository- The World Bank data repository which is based on World Development Indicator (WDI), 2015. Taking the significant factors into account, we build multiple stepwise linear regression models as base models and evaluate model performances using MAE, RAE and RMSE under the shed of k-fold cross validation. Various classifying techniques are observed later using the same data set and the performance of the base prediction model yields more significant.