Abstract

The Multivariate Adaptive Regression Splines (MARS) approach is a multivariate nonparametric regression analysis that assumes the form of a functional relationship between response variable and predictors whose patterns are unknown. MARS is a combination of Recursive Partitioning Regression (RPR) and the Spline method which is able to process high-dimensional data (data that has many predictor variables 3 ≤ p ≤ 20), and large data samples (50 ≤ n ≤ 1000). The MARS model is obtained from a combination of Basis Function values (BF), Maximum Interaction (MI), and Minimum Observation (MO) by trial and error. In this study, we describe the use of MARS to analyze the factors that influence the number of dengue cases in West Sumatra Province. The response variable (Y) used was the number of dengue fever cases in West Sumatra Province with several predictor variables, namely the number of health workers (X1), the number of health facilities (X2), the height of an area (X3), and the density of settlements (X4). The data used is secondary data from the Central Statistics Agency (BPS) in 2019. Based on the calculation, the best model for this problem is the model with a combination of BF = 16, MI = 3, and MO = 2 with GCV value = 279.1654. The results showed that all predictor variables had an effect on the number of dengue fever cases according to the order of importance: the number of health workers (X1), the number of health facilities (X2), the density of settlements (X4) and finally the height of an area (X3).

Original languageEnglish
Article number012094
JournalJournal of Physics: Conference Series
Volume1722
Issue number1
DOIs
Publication statusPublished - 7 Jan 2021
Event10th International Conference and Workshop on High Dimensional Data Analysis, ICW-HDDA 2020 - Sanur-Bali, Indonesia
Duration: 12 Oct 202015 Oct 2020

Fingerprint

Dive into the research topics of 'Factors affecting the number of dengue fever cases in West Sumatra province using the Multivariate Adaptive Regression Splines (MARS) approach'. Together they form a unique fingerprint.

Cite this