Constructing Portfolios According to Financial Statement Information and Copula-GARCH Model in Taiwan Stock Market
碩士 === 國立中山大學 === 財務管理學系研究所 === 102 === Stock selection always has been a challenging and important task. This line of research is highly contingent upon reliable stock ranking for successful portfolio construction. Thanks to recent advances in financial engineering and data mining, we can solve the...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2014
|
Online Access: | http://ndltd.ncl.edu.tw/handle/8n943q |
id |
ndltd-TW-102NSYS5305003 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-102NSYS53050032019-05-15T21:32:36Z http://ndltd.ncl.edu.tw/handle/8n943q Constructing Portfolios According to Financial Statement Information and Copula-GARCH Model in Taiwan Stock Market 以財務報表資訊與Copula-GARCH模型建構投資組合-應用在台灣股票市場 Po-Cheng Lai 賴柏成 碩士 國立中山大學 財務管理學系研究所 102 Stock selection always has been a challenging and important task. This line of research is highly contingent upon reliable stock ranking for successful portfolio construction. Thanks to recent advances in financial engineering and data mining, we can solve these problems more effectively. In this study, we use financial statement information and the ASKSR proposed by Zakamouline and Koekebakker(2009) for stock selection. Furthermore, we apply Copula-GARCH model on Monte Carlo method to generate dynamic optimal weights based on expected utility function. According to this process, we try to construct a quantitative stock selection model which can consistently beat the market. We take all the listed companies in the Taiwan stock market over the period of 1998-2012 as our sample and examine the profitability of portfolios constructed by the combination of different length of in sample and out sample data. The empirical result shows that our portfolios can earn significantly higher return than the TAIEX Total Return Index, especially those that generate optimal weights by mean-variance utility function andγ=1 CRRA utility function. Moreover, after we applied market neutral strategy, we can significantly improve the stability of our portfolios and reduce the possibility of severe losses by the impact of financial crisis. Finally, a robustness test was built to validate if this method works well all the time. We divide our investment period into three shorter periods, and it turns out this method still have great performances on each of the shorter period. As a result, this portfolio construction strategy indeed can be applied consistently and effectively in Taiwan stock market. Chou-Wen Wang Jen-Tsung Huang 王昭文 黃振聰 2014 學位論文 ; thesis 65 zh-TW |
collection |
NDLTD |
language |
zh-TW |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 國立中山大學 === 財務管理學系研究所 === 102 === Stock selection always has been a challenging and important task. This line of research is highly contingent upon reliable stock ranking for successful portfolio construction. Thanks to recent advances in financial engineering and data mining, we can solve these problems more effectively. In this study, we use financial statement information and the ASKSR proposed by Zakamouline and Koekebakker(2009) for stock selection. Furthermore, we apply Copula-GARCH model on Monte Carlo method to generate dynamic optimal weights based on expected utility function. According to this process, we try to construct a quantitative stock selection model which can consistently beat the market.
We take all the listed companies in the Taiwan stock market over the period of 1998-2012 as our sample and examine the profitability of portfolios constructed by the combination of different length of in sample and out sample data. The empirical result shows that our portfolios can earn significantly higher return than the TAIEX Total Return Index, especially those that generate optimal weights by mean-variance utility function andγ=1 CRRA utility function. Moreover, after we applied market neutral strategy, we can significantly improve the stability of our portfolios and reduce the possibility of severe losses by the impact of financial crisis. Finally, a robustness test was built to validate if this method works well all the time. We divide our investment period into three shorter periods, and it turns out this method still have great performances on each of the shorter period. As a result, this portfolio construction strategy indeed can be applied consistently and effectively in Taiwan stock market.
|
author2 |
Chou-Wen Wang |
author_facet |
Chou-Wen Wang Po-Cheng Lai 賴柏成 |
author |
Po-Cheng Lai 賴柏成 |
spellingShingle |
Po-Cheng Lai 賴柏成 Constructing Portfolios According to Financial Statement Information and Copula-GARCH Model in Taiwan Stock Market |
author_sort |
Po-Cheng Lai |
title |
Constructing Portfolios According to Financial Statement Information and Copula-GARCH Model in Taiwan Stock Market |
title_short |
Constructing Portfolios According to Financial Statement Information and Copula-GARCH Model in Taiwan Stock Market |
title_full |
Constructing Portfolios According to Financial Statement Information and Copula-GARCH Model in Taiwan Stock Market |
title_fullStr |
Constructing Portfolios According to Financial Statement Information and Copula-GARCH Model in Taiwan Stock Market |
title_full_unstemmed |
Constructing Portfolios According to Financial Statement Information and Copula-GARCH Model in Taiwan Stock Market |
title_sort |
constructing portfolios according to financial statement information and copula-garch model in taiwan stock market |
publishDate |
2014 |
url |
http://ndltd.ncl.edu.tw/handle/8n943q |
work_keys_str_mv |
AT pochenglai constructingportfoliosaccordingtofinancialstatementinformationandcopulagarchmodelintaiwanstockmarket AT làibǎichéng constructingportfoliosaccordingtofinancialstatementinformationandcopulagarchmodelintaiwanstockmarket AT pochenglai yǐcáiwùbàobiǎozīxùnyǔcopulagarchmóxíngjiàngòutóuzīzǔhéyīngyòngzàitáiwāngǔpiàoshìchǎng AT làibǎichéng yǐcáiwùbàobiǎozīxùnyǔcopulagarchmóxíngjiàngòutóuzīzǔhéyīngyòngzàitáiwāngǔpiàoshìchǎng |
_version_ |
1719116216146067456 |