Publications-NSC Projects
Article View/Open
Publication Export
-
Google ScholarTM
NCCU Library
Citation Infomation
Related Publications in TAIR
題名 統合檢定技術性交易法則對台灣股票市場效率性之影響 其他題名 A Unified Test on the Impact of Technical Rules on Taiwan`s Stock Market Efficiency. 作者 李桐豪 貢獻者 銀行學系 關鍵詞 效率性市場假說;隨機優勢;技術性交易法則;靴帶法;股票市場
Efficient market hypothesis;Stochastic dominance;Technical trading rule;Bootstrap method;Stock market日期 1996 上傳時間 2-Sep-2014 09:00:31 (UTC+8) 摘要 本研究首先探討究竟多少股票的投資方可算是分散風險。結果透過反覆抽樣方式與立方樣條配法發現只要4-5家就夠了。因此,本研究在後續實證所使用的20家股票投資組合應足以分散股市的非系統性風險,而使我們能專注於選股策略與技術性交易法則對投資績效的影響。 本研究所探討的主體則是(1)技術性交易法則是否真的能擊敗市場,而否定台灣證券市場的效率性,(2)不同股票選樣方式是否如Sweeney所言有助於技術性操作之結果,以及(3)謝(1994)與李(1994)所觀察到非常態分配的台灣股票報酬率是否對檢定模擬結果有影響。我們以不同的選股方法再配合上技術性交易法則進行模擬操作,結果發現上述的每一個問題答案都是肯定的。 我們的實證結果顯示投資時若能將選股的策略與技術性交易指標配合運用,則其投資績效表現優於市場指數報酬表現的機率將大幅度地提高。此外,投資組合與技術性交易也沒有造成高風險的投資結果。不過,本研究卻也指出了台灣的股票市場似乎不具有效率性。 我們也發現選股法則有助於投資績效的表現,因此投資的期初選股策略應被列為技術性操作股票時的重要工作。另外,本研究使用的靴帶法統計檢定亦發現傳統t-值檢定力之不足。因此,台灣股票報酬率的非常態分配的特質造成統計上不顯著的問題可能很嚴重。 本研究對技術性操作的肯定態度有可能會刺激市場人士去探究以更多不同的方式或組合來獲取投資的最大利益。或許當市場有足夠不同的技術性操作共同運行時,台灣股市的無效率性問題就不會存在了。
This research project first investigates how many stocks constitute a diversified portfolio. Through repetitive sampling and cubic spline fitting, we find that only 4-5 stocks are sufficient to diversify unsystematic risk. As a result, it should be sufficient for us to utilize 20 stocks in studying the impact of technical trading. Our main research effort is to study (1) whether the technical trading rules really "beats" the market, and the whether Taiwan`s stock market is efficient, (2) whether the initial portfolio selection is really important as claimed by Sweeney, and (3) whether the non-normal distribution of stock returns, evidenced by Hsieh (1994) and Lee (1994), is crucial in testing these simulation results. By using various stock selection methods and technical trading rules, we find the answers for the above questions are all affirmative. Our results indicate that if we could combine stock selection with technical trading rules, we can soundly beat the market performance. We also find that such a strategy does not enhance the risk of our portfolio. This result, however, implies that Taiwan`s stock does not conform to the efficient market hypothesis. We also find that the initial stock selection strategy is important to help the investment performance. Also, the bootstrap method is better than the traditional t-statistic test. This implies that the problem of non-normal distribution characteristic of Taiwan`s stock returns could be serious. Our affirmative conclusion of technical trading may stimulate market practitioners to investigate various combinations of technical trading rules to enhance trading profits. Ironically, such an effort may hopefully and eventually result a more efficient stock market.關聯 行政院國家科學委員會
計畫編號NSC85-2415-H004-011資料類型 report dc.contributor 銀行學系 en_US dc.creator (作者) 李桐豪 zh_TW dc.date (日期) 1996 en_US dc.date.accessioned 2-Sep-2014 09:00:31 (UTC+8) - dc.date.available 2-Sep-2014 09:00:31 (UTC+8) - dc.date.issued (上傳時間) 2-Sep-2014 09:00:31 (UTC+8) - dc.identifier.uri (URI) http://nccur.lib.nccu.edu.tw/handle/140.119/69543 - dc.description.abstract (摘要) 本研究首先探討究竟多少股票的投資方可算是分散風險。結果透過反覆抽樣方式與立方樣條配法發現只要4-5家就夠了。因此,本研究在後續實證所使用的20家股票投資組合應足以分散股市的非系統性風險,而使我們能專注於選股策略與技術性交易法則對投資績效的影響。 本研究所探討的主體則是(1)技術性交易法則是否真的能擊敗市場,而否定台灣證券市場的效率性,(2)不同股票選樣方式是否如Sweeney所言有助於技術性操作之結果,以及(3)謝(1994)與李(1994)所觀察到非常態分配的台灣股票報酬率是否對檢定模擬結果有影響。我們以不同的選股方法再配合上技術性交易法則進行模擬操作,結果發現上述的每一個問題答案都是肯定的。 我們的實證結果顯示投資時若能將選股的策略與技術性交易指標配合運用,則其投資績效表現優於市場指數報酬表現的機率將大幅度地提高。此外,投資組合與技術性交易也沒有造成高風險的投資結果。不過,本研究卻也指出了台灣的股票市場似乎不具有效率性。 我們也發現選股法則有助於投資績效的表現,因此投資的期初選股策略應被列為技術性操作股票時的重要工作。另外,本研究使用的靴帶法統計檢定亦發現傳統t-值檢定力之不足。因此,台灣股票報酬率的非常態分配的特質造成統計上不顯著的問題可能很嚴重。 本研究對技術性操作的肯定態度有可能會刺激市場人士去探究以更多不同的方式或組合來獲取投資的最大利益。或許當市場有足夠不同的技術性操作共同運行時,台灣股市的無效率性問題就不會存在了。 en_US dc.description.abstract (摘要) This research project first investigates how many stocks constitute a diversified portfolio. Through repetitive sampling and cubic spline fitting, we find that only 4-5 stocks are sufficient to diversify unsystematic risk. As a result, it should be sufficient for us to utilize 20 stocks in studying the impact of technical trading. Our main research effort is to study (1) whether the technical trading rules really "beats" the market, and the whether Taiwan`s stock market is efficient, (2) whether the initial portfolio selection is really important as claimed by Sweeney, and (3) whether the non-normal distribution of stock returns, evidenced by Hsieh (1994) and Lee (1994), is crucial in testing these simulation results. By using various stock selection methods and technical trading rules, we find the answers for the above questions are all affirmative. Our results indicate that if we could combine stock selection with technical trading rules, we can soundly beat the market performance. We also find that such a strategy does not enhance the risk of our portfolio. This result, however, implies that Taiwan`s stock does not conform to the efficient market hypothesis. We also find that the initial stock selection strategy is important to help the investment performance. Also, the bootstrap method is better than the traditional t-statistic test. This implies that the problem of non-normal distribution characteristic of Taiwan`s stock returns could be serious. Our affirmative conclusion of technical trading may stimulate market practitioners to investigate various combinations of technical trading rules to enhance trading profits. Ironically, such an effort may hopefully and eventually result a more efficient stock market. en_US dc.format.extent 491 bytes - dc.format.mimetype text/html - dc.language.iso en_US - dc.relation (關聯) 行政院國家科學委員會 en_US dc.relation (關聯) 計畫編號NSC85-2415-H004-011 en_US dc.subject (關鍵詞) 效率性市場假說;隨機優勢;技術性交易法則;靴帶法;股票市場 en_US dc.subject (關鍵詞) Efficient market hypothesis;Stochastic dominance;Technical trading rule;Bootstrap method;Stock market en_US dc.title (題名) 統合檢定技術性交易法則對台灣股票市場效率性之影響 zh_TW dc.title.alternative (其他題名) A Unified Test on the Impact of Technical Rules on Taiwan`s Stock Market Efficiency. en_US dc.type (資料類型) report en
