Analysis of Penalized Semiparametric Regression Model on Bi-Response Longitudinal Data

DOI

10.2991/assehr.k.201010.025How to use a DOI?

Keywords

semiparametric, penalized spline, longitudinal, bi-response, criminality

Abstract

Semiparametric regression is a combination of parametric regression and nonparametric regression. Parametric regression analysis is used if a regression curve or function is known, whereas nonparametric regression analysis is used if the curve form or the regression function is unknown. One of the short descriptions of nonparametric regression analysis is the penalized spline. The penalized spline is a segmented polynomial piece where the data characteristic is explained by knots. The advantage of the penalized approach is flexible and able to describe changes in the behavior patterns of functions within a specific subinterval. In addition, the penalized spline approach can be used to cope with or reduce data patterns that experience a sharp increase. This paper explores semiparametric regression method of the penalized spline by using the longitudinal data of bi-response. The advantage in longitudinal data usage is that it can reduce intervariable collinearity so as to produce an efficient estimate. For case study, we use criminal case data in Indonesia. Based on the results of research, the estimation of penalized spline regression model for bi-response longitudinal data is obtained. Then, the estimation of the penalized regression model is applied in case of criminality and obtained regression model with R-square value of 83.18%.

Copyright

© 2020, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Kosmaryati
AU  - Mujiati Dwi Kartikasari
PY  - 2020
DA  - 2020/10/11
TI  - Analysis of Penalized Semiparametric Regression Model on Bi-Response Longitudinal Data
BT  - Proceedings of the 2nd International Seminar on Science and Technology (ISSTEC 2019)
PB  - Atlantis Press
SP  - 172
EP  - 178
SN  - 2352-5398
UR  - https://doi.org/10.2991/assehr.k.201010.025
DO  - 10.2991/assehr.k.201010.025
ID  - 2020
ER  -