TY - GEN
T1 - Comparison of Multivariate Adaptive Poisson Regression Spline and Multivariate Adaptive Generalized Poisson Regression Spline
AU - Yasmirullah, Septia Devi Prihastuti
AU - Otok, Bambang Widjanarko
AU - Purnomo, Jerry Dwi Trijoyo
AU - Prastyo, Dedy Dwi
N1 - Publisher Copyright:
© 2023 American Institute of Physics Inc.. All rights reserved.
PY - 2023/5/19
Y1 - 2023/5/19
N2 - The multivariate adaptive regression (MARS) spline is a statistical method introduced by Friedman in 1991. The MARS method has several advantages, especially for data that consider the potential nonlinear pattern. These produce a continuous model on knots where the knot selection process is performed via adaptive procedures, capturing additive effects and interactions between predictor variables, and applied to data modeling with numerical and categorical responses. However, the analyzed data is not only numerical or categorical data, but also count type data. Count type data is often found in the health sector, for example, the number of data on tuberculosis. It invites to modify the MARS model with Poisson regression and generalized Poisson regression, i.e. Multivariate Adaptive Poisson Regression Spline (MAPRS) and Multivariate Adaptive Generalized Poisson Regression Spline (MAGPRS). In addition, this study aimed to compare MAPRS and MAGPRS when applied to tuberculosis data. By applying MAPRS and MAGPRS to model the number of tuberculosis in Lamongan regency, the MAPRS method outperforms the MAGPRS method. The MAPRS method has a lower RMSE value than the MAGPRS method.
AB - The multivariate adaptive regression (MARS) spline is a statistical method introduced by Friedman in 1991. The MARS method has several advantages, especially for data that consider the potential nonlinear pattern. These produce a continuous model on knots where the knot selection process is performed via adaptive procedures, capturing additive effects and interactions between predictor variables, and applied to data modeling with numerical and categorical responses. However, the analyzed data is not only numerical or categorical data, but also count type data. Count type data is often found in the health sector, for example, the number of data on tuberculosis. It invites to modify the MARS model with Poisson regression and generalized Poisson regression, i.e. Multivariate Adaptive Poisson Regression Spline (MAPRS) and Multivariate Adaptive Generalized Poisson Regression Spline (MAGPRS). In addition, this study aimed to compare MAPRS and MAGPRS when applied to tuberculosis data. By applying MAPRS and MAGPRS to model the number of tuberculosis in Lamongan regency, the MAPRS method outperforms the MAGPRS method. The MAPRS method has a lower RMSE value than the MAGPRS method.
UR - http://www.scopus.com/inward/record.url?scp=85161417436&partnerID=8YFLogxK
U2 - 10.1063/5.0119385
DO - 10.1063/5.0119385
M3 - Conference contribution
AN - SCOPUS:85161417436
T3 - AIP Conference Proceedings
BT - Proceedings of the International Conference on Advanced Technology and Multidiscipline, ICATAM 2021
A2 - Widiyanti, Prihartini
A2 - Jiwanti, Prastika Krisma
A2 - Prihandana, Gunawan Setia
A2 - Ningrum, Ratih Ardiati
A2 - Prastio, Rizki Putra
A2 - Setiadi, Herlambang
A2 - Rizki, Intan Nurul
PB - American Institute of Physics Inc.
T2 - 1st International Conference on Advanced Technology and Multidiscipline: Advanced Technology and Multidisciplinary Prospective Towards Bright Future, ICATAM 2021
Y2 - 13 October 2021 through 14 October 2021
ER -
