Predicting the Number of Children Ever Born Using Logistic Regression Model
Md. Tofazzal Hossain · Biometrics & Biostatistics International Journal · 2015
Modeling total children ever born is very crucial since Bangladesh is going under a rapid growth of population.We use logistic regression model, a well-known technique for categorical data analysis, to predict the number of children ever born using BDHS (2011) data.BDHS data is not random, since two stages stratified sampling is conducted to collect the data.Cluster correction and weight adjustment is necessary for analyzing this data.An extensive literature search fails to find any work which uses these adjustments for predicting the number of children ever born.In this paper these two adjustments have been used for predicting the number of children ever born.The prediction performance has been measured by training and test error.