Application of Groupwise Principal Sensitivity Components on Unbalanced Panel Data Regression Model for Gross Regional Domestic Product in Kalimantan

Most robust estimation methods for panel data regression models do not consider the panel data structure consisting of several cross-sections and time-series units. This robust method, which does not consider the panel data structure, can completely remove all observations from a cross-section unit...

全面介紹

Saved in:
書目詳細資料
Main Authors: Yuniarti, Desi, Rosadi, Dedi, Abdurakhman, Abdurakhman
格式: Article PeerReviewed
語言:English
出版: Universiti Putra Malaysia Press 2022
主題:
在線閱讀:https://repository.ugm.ac.id/278688/1/Yuniarti_MA.pdf
https://repository.ugm.ac.id/278688/
http://www.pertanika.upm.edu.my/
https://doi.org/10.47836/pjst.30.4.01
標簽: 添加標簽
沒有標簽, 成為第一個標記此記錄!
實物特徵
總結:Most robust estimation methods for panel data regression models do not consider the panel data structure consisting of several cross-sections and time-series units. This robust method, which does not consider the panel data structure, can completely remove all observations from a cross-section unit in trimming outlier observations. However, it can cause biased estimation results for the cross-section unit. This study determines the robust estimate for the unbalanced panel data regression model using Groupwise Principal Sensitivity Components (GPSC) by considering grouped structure data. The results were compared with Within-Group (WG) estimation and other robust estimation methods, namely Within-Group estimation with median centering (Median WG), Within-Group Least Trimmed Squares (WG-LTS), and Within Generalized M (WGM) estimators. Comparisons were made based on the Mean Squares Error (MSE) value. In this study, we applied the proposed method to the unemployed and the Gross Regional Domestic Product (GRDP) data at constant prices in Kalimantan, Indonesia. The analysis showed that GPSC was the best method with the smallest MSE value. Therefore, we can consider implementing and developing the GPSC method to detect and determine the robust estimates for the unbalanced panel data regression model because it fits the panel data structure.