Parameter Estimation Robust Regression Method of Moment (MM) in Cases of Maternal Death in Indonesia

Abstract

Regression analysis is used to determine the relationship between the dependent and
independent variables with a parameter estimator. The parameter estimator that is
usually used is the Least Squares Method (LSM), this requires a classical assumption
test. Some cases have normality assumptions that are unfulfilled because there are
outliers so the result regression parameter estimates are not accurate so that robust
regression is used in the analysis. Robust regression is a regression analysis method that
can withstand outliers. The purpose of this study is the application of robust regression
estimation Method of Moment (MM) with Tukey Bisquare weighting in the case of data
on the number of maternal deaths in Indonesia 2020 with the number of maternal deaths
as a dependent variable, and with independent variables such as the number of pregnant
women who experience bleeding, the number of diabetics in pregnancy, and the number
of HIV positive in pregnancy. The result showed that every one unit increase of three
independent variables had a positive effect on the number of cases of maternal deaths,
each of which was 2,8064; 2,5014; 1,1577.