投資人行為與學習效果之研究:以處分效果與過度自信為例

No Thumbnail Available

Date

2018

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

本研究旨在探討兩種投資人其行為偏誤和學習效果間之關聯性,並以臺股期貨交易市場為例,篩選出具不同程度處分效果與過度自信之投資人,各自再分成兩種不同交易型態包含累積交易量高且累積交易天數長以及累積交易量低且累積交易天數短之投資人,觀察市場上兩類具行為偏誤之投資人學習效果情形。我們分別利用累積交易量及累積交易天數作為經驗的衡量指標,試圖檢驗投資人隨經驗累積是否能從中獲得學習能力,並透過所產生的學習效果降低其行為偏誤程度。我們發現若用累積交易量來衡量經驗時,對於累積交易量低且累積交易天數短之處分效果投資人,交易量與其行為偏誤程度大致呈負向關係,表示投資人確實能透過交易量累積來產生學習效果並改善行為偏誤程度。而對於累積交易量高且累積交易天數長之過度自信投資人,交易量與其行為偏誤程度呈正向關係,代表投資人存在自我歸因偏誤而使過度自信程度增加,透過交易經驗累積而產生學習成為過度自信之現象。此外,若改用累積交易天數作為經驗的衡量指標時,我們觀察到無論是對於何種交易型態之處分效果或過度自信投資人,交易天數與其行為偏誤程度近乎呈現顯著負相關,顯示投資人可經由其累積交易天數的增加,並透過學習行為來增進自身交易能力,以降低其行為偏誤程度。
This study aims to examine the relation between nonrational behavior and learning effect in two types of investors. We analyze investors with different degrees of disposition effect and overconfidence in Taiwan futures market, dividing into two trading patterns including investors with high accumulated trading volume and long accumulated trading days and investors with low accumulated trading volume and short accumulated trading days in order to observe the learning effect of investors with nonrational behavior in different trading patterns. Further, we employ accumulated trading volume and trading days respectively as a proxy for experience because of bifurcated conclusions. The current paper investigates whether the degree of investors’ nonrational behavior will decline by learning about their ability through experience. Interestingly, if using trading volume as a measure of experience, we present evidence that there is a negative relationship for disposition effect between trading volume and the degree of nonrational behavior for those investors with low accumulated trading volume and short accumulated trading days. Indeed, it shows that investors’ learning behavior do improve their nonrational behavior; however, there is a significantly positive relationship for overconfidence between trading volume and the degree of nonrational behavior for those investors with high accumulated trading volume and long accumulated trading days, indicating there’s a self-attribution bias existing in investors with overconfidence, which makes them tend to have more propensity for learning to be overconfident with increasing trading experience. Moreover, if trading days is used to proxy for experience, there is no difference between disposition effect and overconfidence by focusing on the nonrational behavior and trading patterns. Both demonstrate negative correlation between trading volume and the degree of nonrational behavior, which concludes that investors acquire better trading ability through learning behavior with accumulation of trading days, improving their nonrational behavior.

Description

Keywords

處分效果, 過度自信, 學習效果, 固定效果模型, Disposition Effect, Overconfidence, Learning Effect, Fixed Effect Model

Citation

Collections

Endorsement

Review

Supplemented By

Referenced By