Abstract Due to the arrival of global zoned economy, the fluent fund has promoted the intercommunication of international politics and economy. The multiple investments have become research focus in recent years, leading to the reasons of the relationship and fluctuation in various countries. Thus this research, taking Taiwan as a starting point, analyzes the relationship of stock price with one of our neighbor countries, South Korea, including their deep bid, four major type stocks ( plastics, transportation, steel, electronics), nine major personal shares( Taiwan Plastics and S.K. Chemical Industry, Evergreen Shipping and Han Jin Shipping Co., China Steel Co. and Posco, Taiwan Semiconductor Manufacturing Co. and Samsung Electronics, AUO and Samsung Electronics). By use of E-view software, it analyzes their closing price from June 2001 till June 2005, expecting to improve the investment performances of the government, investors and relevant industries. The study shows that every stock price index has arrays of single root. Going on with Johansen Cointegration relationship with Granger Causality, it can get the following results: I) Only short term relation exists for deep bid, other three major type stocks and nine main industries¡¦ personal shares between S. Korea and Taiwan, except electronic stocks. II) No short term relation exists among deep bid, electronic type stock, Evergreen and Han Chin Shipping Co. personal shares. As for plastics, transportation, steel type stock, China and Posco, Taiwan Electronics and Samsung¡¦s personal shares, S. Korea is a leading indicator as it influences Taiwan¡¦s present situation. On the other hand, Taiwan influences S. Korea as a leading indicator for Taiwan Plastics and S.K. Chemical Industry. These mentioned above are mono-way cause and effect. Finally, research even shows that there¡¦s a mutual cause and effect relation between AUO and Samsung Electronics. III) Taking a general view of weigh value of equity market and short term relation, it can sum up to the following results : 1) The results are the same comparing the leading and backward relation of type stocks to large proportion personal shares. 2) The leading and backward relation of small proportion does not have the same influence as type stocks. 3) The type stocks, that originally do not have cause and effect relation, will appear mutual influence relation if the personal shares take large proportion in type stocks, due to high similarity of production among industries. Key Words: Taiwan stock market , South Korea stock market, Cointegration relationship.

This study mainly investgates the investment performance of momentum strategies and contrarian strategies in Taiwan stock market. There are three purposes in this paper. First, we examine whether the momentum strategies and contrarian strategies can create significant profits under different formulation horizons and holding horizons, then we discuss the reasons for the profits of significant profits strategies, including risk, seasonality, industrial momentum, time series predictability of stock returns and cross-sectional variation in the mean returns, and stock underreation, overreaction, and random walk. Second, we derive the mix strategies from the combination of momentum strategies and contrarian strategies for the same holding horizons and test the investment performance of mix strategies empirically. Finally, we study whether the investment strategies of stock mutual funds in Taiwan are industrial momentum strategies or industrial contrarian strategies, and which strategies can create better industrial investment performance. The main conclusions and suggestions are as follows: First, we find the momentum strategies are more successful in Taiwan as a whole, especially from 1991/1/1 to 2000/12/31. There are only three significant profits strategies in 147 strategies totally for three different test periods, including the (24,24) strategy and (36,24) strategy from 1991/1/1 to 2000/12/31, and (1,12) strategy from 1981/1/1 to 1990/12/31. For the reasons of the profits of the three strategies, we find the negative alphas in the F&F three factors model and underreation from the decreasing returns in the post holding horizons, so we should use the momentum strategies very carefully in Taiwan stock market. Second, we find the success of mix strategies theoretically and empirically. Nevertheless, we can¡¦t increase the profits for considering more different sub-strategies if there are no successful sub-strategies with different formulation horizons. Finally, we find the investment strategies of stock mutual funds almost are industrial momentum strategies, which realized significantly better industrial performance then the industrial contrarian strategies. It suggests that the industrial momentum strategies are not irrational and can increase the speed of adjustment of industrial index to its intrinsic value. On the other hand, stock mutual funds can perform well by the momentum strategies without superior information collection and analysis.

Dividend Valuation Model is a well-known stock pricing model. However, many empirical studies of foreign stock markets do not support the Dividend Valuation Model; most of these studies think stock price is too volatile to explain by expected dividend. Therefore, this article would like to use Stochastic Cointegration to reexamining Taiwan stock market, and observe whether Taiwan stock market supports Dividend Valuation Model. The empirical results showed that stock price and dividends exist a positive comovements relationship in the plastic, steel, electronic, and the banking & insurance industries, but empirical results does not completely support the theoretical value of cointegration vector. Therefore, this study has not been sufficient evidence to support Taiwan stock market is efficient.

金融資產報酬的波動一直都是財務市場熱衷研究的主題, 由於真正報酬的波動無法確知, 造成無法判斷何者為衡量報酬波動最佳的模型, 進而導致預測未來報酬的風險增加。因此, 本文利用靴帶抽樣法(Bootstrap)反覆抽樣的估計方式, 建立報酬與報酬波動的預測區間來衡量由估計模型參數產生的不確定性, 希望能藉此更瞭解資產報酬的變化以降低投資風險。鑒於目前衡量報酬波動的模型眾多, 文中將採用文獻上普遍最能掌握金融資產報酬波動現象的GARCH模型, 作為衡量報酬波動的方法, 再以靴帶抽樣方法估計其報酬與報酬波動的預測區間, 透過有限樣本的模擬將估計模型參數不確定性的靴帶抽樣方法與其他方法比較, 證明靴帶抽樣法最能適當的捕捉報酬波動真實的情況。最後, 由台灣上市股票市場中選取四支不同類股的各股以日報酬進行實證研究, 結果顯示各股的日報酬都具有波動變異的現象, 進一步估計樣本外不同範圍的波動預測區間, 發現利用估計模型參數不確定性的靴帶抽樣方法可以適當地涵蓋波動的變化。

台灣在2008年二次政黨輪替後，提出若干優惠措施以及取消某些法規限制，積極推動台商回台上市，然而深圳證券交易所在2009年發展出創業板，其發展目的是為了讓新興企業有一個高效的籌資平台，深圳創業板的開板給了台商在中國大陸上市的契機與吸引力，因此台商回台上市的意願便可能會因深圳創業板的開板而降低。本研究將以台商的觀點來看，當台商面臨回台上市或在深圳創業板上市兩者之間做選擇時，對於兩者制度面、市場面及環境面來進行比較分析，並藉此來判斷深圳創業板的發展是否會影響台商回台上市的意願。 研究中發現，目前深圳創業板吸引台商之處在於其上市門檻較低，本益比及活絡性表現較好，再加上中國大陸的經濟環境較佳，因此可募集到較多的資金。雖然台灣證券市場現階段在各方面的成長幅度不如深圳創業板突飛猛進，但台灣證券市場是一個較為成熟的證券市場，其多年累積下來的國際性、穩定度及成熟度皆比深圳創業板優越許多。因此，台灣證券市場目前的整體表現仍優於深圳創業板，深圳創業板的開板尚不至於影響台商回台上市的意願。 最後建議未來台灣證券市場應積極爭取與兩岸三地證券市場更密切的接觸，以「區域整合」作為當前的發展方向，若能合作組成「大中華區域證券市場」─雙邊掛牌機制甚或多邊掛牌，以兩岸三地之間的優勢互補，更能達到競合、更甚是綜效的效果。 / After the second return of the governing party in Taiwan, the new government released some beneficial programs and deregulated some laws to attract Taiwanese enterprises to IPO in Taiwan. In the meanwhile, ChiNext opened in 2009 for the purpose of offering an efficient capital-raising platform for those newly-formed enterprises. ChiNext gives Taiwanese enterprises an attractive opportunity to IPO in China, and therefore Taiwanese enterprises may decrease their willing to IPO in Taiwan. This paper starts from the opinion of Taiwanese enterprises who make a decision to IPO in Taiwan or ChiNext, analyzing by the side of regulation, market and environment to determine whether ChiNext will affect the willing of Taiwanese enterprises to IPO in Taiwan or not. What attracts Taiwanese enterprises to IPO in ChiNext is its excellent PE ratio, high turnover rate and being easy to be qualified. In addition, owing to the rapid growth in Chinese economy, enterprises will raise much more capital in ChiNext. Contrary to ChiNext, Taiwan stock market is such a mature stock market that it could not expand rapidly as ChiNex. Also because of its maturity, Taiwan stock market is much steadier and much more international than ChiNext. To sum up, Taiwan stock market is still more outstanding than ChiNext, the appearance of ChiNext does not affect the willing of Taiwanese enterprises to IPO in Taiwan at present. Finally, this paper makes a suggestion of "region conformity" as a direction for Taiwan government. Taiwan government should connect with Chinese-area stock market more intensely. If the stock market of Taiwan, Hong Kong and China could form a "Chinese-area Stock Exchange" like NYSE Euronext or other else, it will create an effect of synergy in Chinese-area stock market.

本研究最主要的貢獻為建構一具台灣股市投資人情緒指數並檢測投資人情緒指標與台灣股市的關係。本研究以台灣股票市場為背景，研究期間為2001年1月至2010年12月。利用Baker, Wurgler and Yuan在2009年提出的方法以Volatility Premium, Number of IPOs, First Day Return of IPOs, Turnover Rate四個變數編製台灣股市投資人情緒指數，並探討台灣股市投資人情緒指數變動量與台股大盤報酬之間的領先落後關係。 實證結果發現，在較短的時間，如月資料，台股大盤報酬會影響下一期的台灣股市投資人情緒指數變動量，而在較長的時間，如季資料，台灣股市投資人情緒指數變動量會影響四期後的台股大盤報酬，即短期台灣股市投資人情緒指數變動量為大盤報酬之落後指標，長期則為大盤報酬之領先指標，短期原因為投資人情緒指數受大盤報酬影響，而易有追高殺低現象，長期雖投資人情緒領先大盤報酬的結果在統計上顯著，但經濟上並無顯著意義，另一方面，台股大盤報酬與台灣股市投資人情緒指數變動量間存在正相關，即不能以台灣股市投資人情緒指數變動量預測股市泡沫。 / The main contribution would be the construction of the sentiment index in Taiwan stock markets and examining the relationship between the variation of the sentiment index and Taiwan stock market returns. The background is Taiwan stock markets. The sample period is from January 2001 to December 2010. We use the method in Baker, Wurgler and Yuan (2009) to measure investors’ sentiment and explore the relationship between the variation of the sentiment index and Taiwan stock market returns. The empirical result reveals that in monthly data, Taiwan stock market returns is the leading indicator of the variation of investment sentiment. In a longer term, we mean the quarterly data in this paper, the situation changes. In quarterly data, the variation of the investment sentiment is the leading indicator of the Taiwan stock market returns. In addition, instead of a negative correlation between the stock market returns and our sentiment index, we prove that our sentiment index have a positive impact on stock market returns. Therefore, we could not use this sentiment index to forecast future economic bubbles.

本研究利用479個全球指標對台指期建立預測模型。該模型可以預測台指期在未來K天的漲跌走勢。我們使用了兩種演算法（支持向量機和決策樹）以及兩種取樣方式（交叉驗證和移動視窗）進行預測。在交叉驗證的建模過程中，決策樹展現了較高的預測力，最高準確度達到了93.4%。在移動視窗的建模過程中，支持向量機表現較好，達到了79.97%的預測准確度。於此同時，不管是哪一種條件設定都表明當我們預測的週期拉長時，預測的效果相對較好。這說明全球市場對台灣市場的影響很大，但是需要一定的市場反應時間。該研究結果對投資人有一定的參考作用。在未來方向裡，可以嘗試使用改進的決策樹演算法，也可以結合回歸預測進行深入研究。 / In this research, we build a stock price direction forecasting model with Taiwan Stock Index Future (TXF). The input data we used is 479 global indices. The classification algorithms we used are SVM and Decision Tree. This model can predict the up and down trend in the next k days. In the model building process, both cross validation and moving window are taking into account. As for the time period, both short term prediction (i.e. 1 day) and long term prediction (i.e. 100 days) are tested for comparison. The results showed that cross validation performs best with 93.4% in precision, and moving window reached 79.97% in precision when we use the last 60 days historical data to predict the up and down trend in the next 20 days. The results imply Taiwan stock market is significantly influenced by the global market in the long run. This finding could be further used by investors and also be studied with regression algorithms as a combination model to enhance its performance.

摘要 2007年初美國發生次級房貸大量違約, 陸續有銀行倒閉, 進而撼 動整個美國與歐洲股市。一向與美國有密切貿易關係的台灣,在此事 件中到底受到多大的影響? 本文利用DCC模型探討次貸風暴前後,台 美股價間的關係是否有發生顯著的變化? 實證結果發現: 台灣與美國 的動態相關係數在次級房貸之後, 反而變小, 可見台灣的股市並未受 到很大的衝擊, 而亞洲地區的大多數國家也都與台灣相似,與美國的 動態相關係數變小,可見亞洲地區在次貸風暴中扮演著避風港的角色。

本文主要在探討在何種情況下(意即金融危機發生前後)會發生比較顯著的跨國從眾現象。本文採用2003年10月1日到2009年2月28日期間的台灣加權指數、台灣50、美國S&P 500和道瓊工業指數報酬率資料，並利用Hwang and Salmon (2004)測量從眾行為指標的方法，檢驗在此段期間下，台灣投資人行為是否有明顯跟隨美國投資人行為的現象。實證研究發現整體而言在此段期間下，台灣存在顯著的跨國從眾行為。細部以月為單位探討從眾行為顯著結果的變化，可以歸納出以下三個結論：一、從眾行為主要發生在經濟情況相對穩定的情況下，意即在金融危機日趨嚴重以前，測量從眾行為的指標反而比較顯著。例如在2007年和2008年時的經濟情況比在2003年、2004年和2005年時還要衰退，但測量到存在顯著跨國從眾行為的月數反而較少。二、持續存在顯著跨國從眾行為的最長期間為2005年3月到2006年1月，歷經11個月。三、最常被觀測到有顯著從眾行為現象的月份為1月、11月和12月。 / This paper mainly examines under what conditions herding behavior is likely to become more significant and obvious, in which I modify Hwang and Salmon (2004)’s methodology and use the returns data of Taiwan Weighted Index, Taiwan 50 stock Index, S&P 500 stock and Dow Jones Industry Index of the sample period of October 1, 2003 to February 28, 2009 to test if there is any multinational herding behavior. I find that Taiwan investors in this sample period follow (even imitate) the investment actions of American investors. In more details about the herding patterns, we have found three main phenomena. First, herding behavior mainly occurs significantly during relatively quiet period, say, before the financial crisis, rather than when the market is under stress. The economic situations in 2007 and 2008 are much worse than in 2003, 2004, and 2005, but numbers of months exhibiting significant herding in these bad situations are less. Second, the longest lasting time of herding is March 2005 to January 2006, which totally lasts for eleven months, and this period is before 2007 and 2008 in which the financial systems are destroyed badly. Third, herding always happens in January, November and December given the sample period.

在國內外實證文獻中，已有許多學者研究市場過度反應的現象，但由採取的研究方法不盡相同，故結論也有所差異．本研究是希望能對台灣股市的個別股票報酬率作一較完整的檢視，先瞭解股票報酬率前後期的相關情形，試圖找出一些一致性，再提出股市交易的策略，並比較策略的獲利性。 將樣本分成對稱與不對稱的形成期及檢定期，採Spearman等級相關係數法先對台灣股市同一股票的報酬率在不同期間（形成期及檢定期）的表現是否有相關作一檢視，接著檢定此相關係數時間序列是否具有隨機的特質，而後將證券交易稅及手續費納入考慮後，比較三種投資策略－買入持有輸家、中間、贏家的獲利性、風險及績效表現。 本文的實證結果：由單位風險報酬率來看，投資股市應以買入有長期的策略才會有較好的績效表現，而買入持有短期的績效是最差的。以不同的樣本期間討論台灣股市是否有價格反彈，結果並不相同。在1980年-1998年的Spearman等級相關係數多為正值，表示市場在288天以下多沒有價格反彈，且以連檢定的結果多為顯著，以Jensen α檢定之，贏家投資策略能獲得超額報酬。在1990年-1998年不論是重覆取樣、未重覆取樣，Spearman等級相關係數，多為負值，表示市場應存有價格反彈的現象，且以連檢定的結果多為顯著，但以Jensen α檢定之，輸家卻無法獲得超額報酬。可能是因為考慮了交易稅與手續費，而影響了投資策略的獲利性。 / There have been many articles discussing overreaction. Because of the difference of methods and samples, the conclusions are different. This thesis tries to make a more complete examination of Taiwan Stock Market. We divide sample period to be formation period and test period which are symmetric and asymmetric. Besides that, we adopt overlapping and nonoverlapping sampling. The sample period is 1980 January 1 to 1998 January 22. Three investment strategies are buying and holding loser portfolio、middle portfolio and winner portfolio. We use Spearman rank correlation to discuss whether the return of Taiwan Stock Market has correlation between formation period and test period. Then we adopt one of nonparameter statitics analysis－run test to examine whether the time series of Spearman rank correlation is a random walk. Following are our summaries： 1.The longer period we hold the stock,the better return we acquire. 2.In the first sample (1980 Jan 1 to 1998 Jan 22), Spearman rank correlation. is almost positive, and the hypothesis of run test is significant. We imply buying and holding the winner portfolio is the best strategy. We can prove this by using Jense α. In this case, buying and holding winner can get excess return. 3.In the second sample(1990 Nov 1 to 1998 Jan 22), Spearman rank correlaion is almost negative, and the hypothesis of run test is signficant. We imply buying and holding the loser portfolio is the best strategy. But we can not prove the by using Jense α. As we can not acqure excess return by buying and holding loser portfolio.