Open AccessANALISIS KEMISKINAN DI KABUPATEN MALUKU TENGGARA BARAT MENGGUNAKAN PENDEKATAN MULTIVARIATE ADAPTIVE REGRESSION SPLINE (MARS)
Author(s) -
Ferry Kondo Lembang,
Henry W. M. Patty,
Feros Maitimu
Publication year - 2019
Publication title -
media statistika
Language(s) - English
Resource type - Journals
ISSN - 2477-0647
DOI - 10.14710/medstat.12.2.188-199
Subject(s) - multivariate adaptive regression splines , mars exploration program , toilet , population , poverty , clothing , mathematics , geography , statistics , socioeconomics , regression analysis , toxicology , engineering , environmental health , environmental engineering , economics , biology , bayesian multivariate linear regression , economic growth , medicine , archaeology , astrobiology
Poverty is a condition where there is a condition where there is an inability of the community to meet basic needs such as food, clothing, shelter, education and health. MTB regency is one of the regions in Moluccas Province with a relatively high percentage of the poor population reaching 28.31%. The purpose of this study is to conduct poverty analysis in MTB using the MARS method. The problem of poverty is thought to be very much influenced by many factors, therefore the selection of the MARS method is considered very appropriate because it has the advantage of being able analyze high-dimensional data. The results showed the best MARS model was a combination BF=18, MI=3 and MO=0 with a minimum GCV value at 69.587. Variables that have a significant effect are the percentage RTM that do not have public toilet facilities (X5), the variable percentage of RTM that is the type of floor of a residential building made of poor quality soil / bamboo / wood (X4), and the percentage of RTM that does not own the building (X1).