MODELING OF THE PERCENTAGE OF AIDS SUFFERERS IN EAST JAVA PROVINCE USING THE MULTIPREDICTOR NONPARAMETRIC REGRESSION APPROACH BASED ON SPLINE TRUNCATED ESTIMATOR

Abstract

AIDS or Acquired Immune Deficiency Syndrome is a set of symptoms and infection or a syndrome that arise due to damage to the human immune system. AIDS is a health problem that often occurs in developing countries, including in Indonesia. East Java Province was ranked first in the highest number of AIDS sufferers in Indonesia ever reported from 1987-2016 as many as 16,911 people out of a total of 86,780 people. In order to overcome AIDS cases, it is necessary to know the factors that influence it. Data on the percentage of AIDS sufferers and their predictor variables have irregular data patterns or do not match in certain patterns, then the method that can solve these problems is by using the nonparametric regression based on spline truncated estimator. A spline truncated estimator is a segmented polynomial function that has better flexibility because there are knot points indicating changes in data behaviour patterns. The data that used in this study is a secondary data in 2016 obtained from the East Java Provincial Health Office. The results showed that the R² value generated from the best model was 93.84%. This shows that the variables of health facilities, blood donors, health workers, condom users, and residents of 25-29 years are able to explain 93.84% of the percentage of AIDS sufferers in East Java Province in 2016.