A Study on Estimating Standard Errors using Taiwan Financial Panel Data

• Main Authors: Chien-chung Lin, 林建忠
• Other Authors: Chin-sheng Huang
• Format: Others
• Language: zh-TW
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/23868842953202140912

碩士 === 國立雲林科技大學 === 財務金融系碩士班 === 100 === Financial researchers commonly encounter financial panel data sets. Unwarranted conclusions are frequently resulted from uncorrected estimation methods (e.g., OLS, Fama-MacBeth standard errors) when the residuals are correlated across firms or across time. This study examines a wide array of methods employed in financial panel data sets and simulates the magnitude of bias incurred by those methods when confronting different clustering panel structures. Empirically, this study also experiments the issue on a real Taiwanese financial panel set. Our main finding indicates that the time effect can be severely misleading the inference when the number of time clusters is relatively smaller. This study might sever as a general guidance to assisting researchers how to cope with financial panel data sets.