BOOTSTRAP AGGREGATING (BAGGING) REGRESI LOGISTIK ORDINAL UNTUK MENGKLASIFIKASIKAN STATUS GIZI BALITA DI KABUPATEN KLUNGKUNG

This research was conducted to determine the variables that significantly influence nutritional status of children based on indicators that defined as height for age (H/A) and to classify children nutritional status into normal, short or very short categories. Height for age (H/A) is indicator used to describe the circumstances of malnutrition short. Short children (stunting) is children who fail to reach optimal growth. The secondary data was list of 116 data of children aged 24-59 months at UPT. Puskesmas Klungkung I in 2015. The method was used was ordinal logistic regression and bagging ordinal logistic regression. Based on the research results, it was obtained variables children body length at birth, birth weight, and length of mid-upper arm circumference (MUAC) in pregnant woman were significantly affects the nutritional status of children by the classification accuracy level of ordinal logistic regression and misclassification . Classification accuracy of ordinal logistic regression can be improved by bagging ordinal logistic regression method. Bagging works well on classification method which has unstable procedures. One of classification method which has unstable procedures is ordinal logistic regression. Bagging ordinal logistic regression method by 501 times replication capable to improve classification accuracy of ordinal logistic regression model from to , increased .