Regression for Trend-Seasonal Longitudinal Data Pattern: Linear and Fourier Series Estimator

M. Fariz Fadillah Mardianto, Sri Haryatmi Kartiko, Herni Utami

2018

Abstract

Longitudinal data is a pattern that consists of time series and cross section data pattern. In a research with longitudinal and panel data often be used combination between trend and seasonal or trend-seasonal pattern, for example the relationship between profit and demand for seasonal commodities, in education insurance, meteorology case and many more for many subjects. Recently, we develop Fourier series estimator to approach curve regression for longitudinal data. Fourier series that be used, not only include trigonometric Fourier series which usual be used in Mathematics, but also linear function. In this research we compare performance of new estimator with linear estimator that often be used in panel data regression or parametric regression for longitudinal data. The trend-seasonal data that be used in this analysis is gotten from simulation process based on Box et.al., (1976). The Fourier series estimator gives better result with goodness indicator smaller Mean Square Error (MSE) and greater determination coefficient than linear estimator.

Paper Citation

in Harvard Style

M. Fariz Fadillah Mardianto., Kartiko S. and Utami H. (2018). Regression for Trend-Seasonal Longitudinal Data Pattern: Linear and Fourier Series Estimator.In Proceedings of the International Conference on Mathematics and Islam - Volume 1: ICMIs, ISBN 978-989-758-407-7, pages 350-356. DOI: 10.5220/0008521803500356

in Bibtex Style

@conference{icmis18,
author={M. Fariz Fadillah Mardianto and Sri Haryatmi Kartiko and Herni Utami},
title={Regression for Trend-Seasonal Longitudinal Data Pattern: Linear and Fourier Series Estimator},
booktitle={Proceedings of the International Conference on Mathematics and Islam - Volume 1: ICMIs,},
year={2018},
pages={350-356},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0008521803500356},
isbn={978-989-758-407-7},
}

in EndNote Style

TY - CONF

JO - Proceedings of the International Conference on Mathematics and Islam - Volume 1: ICMIs,
TI - Regression for Trend-Seasonal Longitudinal Data Pattern: Linear and Fourier Series Estimator
SN - 978-989-758-407-7
AU - M. Fariz Fadillah Mardianto.
AU - Kartiko S.
AU - Utami H.
PY - 2018
SP - 350
EP - 356
DO - 10.5220/0008521803500356