How to evaluate mutual fund performance correctly and determine the investment targets of mutual funds are the important issues to investors. In this study, we apply an innovative bootstrap statistical technique, to solve the small sample size problem and the distribution assumption disturbance in previous research. We examine the performance of domestic open-end mutual funds over the period from 1998 to 2003 using five performance measurement models. We further test the persistence of mutual fund performance. This study shows that¡G 1. On average, mutual fund managers do not own superior ability in stock selection. Most funds experiencing abnormal performance may simply result from good luck, since random selection also creates abnormal performance. 2. Mutual fund managers do not own market-timing ability. Classified further by investment objectives, the sample indicates that only the group of small-scale stocks shows significant market-timing ability. 3. Performance persistence does not exist no matter in long-term or short-term period.

I examine mutual fund performance using three different perspectives. I begin with Mutual Fund Holdings Batting Average, in which I analyze mutual fund performance through the creation of a new variable using funds’ stock holdings information. My results show that this new variable, Holdings Batting Average, is related to the future performance of managers. My next chapter, Quarterly Mutual Fund Holdings Information and Window Dressing examines two different approaches of using holdings information. I recommend that fund holdings reported at the beginning of the quarter are more related to actual mutual fund performance than holdings disclosed at the end of the quarter. In my last chapter, Morningstar’s Upside and Downside Capture Ratios¸ I test these two ratios that are being reported by Morningstar. I find that these measures do not predict outperformance, but appear to be related to future fund flows.

在資產管理公司的全球化浪潮下，國內資產管理的規模大幅成長，其商品種類也不斷地推陳出新，而投資人再選取商品上，經由過去的文獻發現投資人自身在判斷一檔基金投資與否通常會看其過去的績效，在近期有較高績效的共同基金較會受到投資人青睞。故在投資人有這種追求過去歷史績效的現象時，如何選擇有績效持續性的共同基金就變成一個重要的議題。 而一直以來，基金績效持續性的探討所找到的結果各家看法不一，部分文獻顯示出基金績效的持續性非來自於基金經理人的強調選股特徵，而亦有學者認為基金經理人可能會有能力上或是訊息上的優勢。經由歸納，常發現已發展國家的股票型基金持續性是不顯著的，而新興國家可能由於經理人的資訊程度較大眾的消息取得容易且迅速。 本文經由探討台灣股票型基金發現擁有較好強調選股特徵(1-R2)的基金其在未來績效較有持續性，若再搭配當期α來考量，則短期，投資強調選股特徵弱但α大的群組或是投資強調選股特徵強但α小的群組，績效表現較好；然而，若放眼長期，擇投資強調選股特徵強的基金，績效表現會較為出色，尤其是α落在較大群族的基金。整體而言，淨資產對於持續性的影響是顯著負向的，可能原因為規模不經濟導致；週轉率越高代表其績效持續性較強。另外新資金湧入導致基金績效持續性較不佳，原因可能為其淨資金流入會造成基金操作管理上效率的問題。 / Under the globalization tide of the asset management company, the asset managed in Taiwan has been grown dramatically, and much more various products have been launched. Empirical evidence found that investors tend to take past performance into consideration before they invest in funds. As a consequence, funds with recent outstanding performance are more popular than others. So performance persistence becomes an important issue. Empirical researches did not reach consensus on whether funds have performance persistence, some paper shows that performance persistence does not stem from stock picking ability, however, some evidence show that fund managers might have some information or ability advantages. And performance persistence is more likely happened in emerging countries than developed countries due to fund managers have more efficient and latest information than general investors. This paper finds that Taiwanese stock fund which emphasize more on stock picking ability (higher 1-R2) tend to persist. If analyzed with current α, funds with less emphasis on stock picking and bigger current α in the short run, or funds with more emphasis and weaker α will have better performance in the future. In the long run, more emphasis on stock picking, better performance in the future，especially those with strongerα. Greater asset under management and net sales rate might cause worse performance persistence due to inefficiency in management. And higher turnover help performance persistence.

Die Auswahl von Aktien ist ein Gebiet der Finanzanalyse, die von speziellem Interesse sowohl für viele professionelle Investoren als auch für Wissenschaftler ist. Empirische Untersuchungen belegen, dass Aktienerträge vorhergesagt werden können. Während verschiedene Modellierungstechniken zur Aktienselektion eingesetzt werden könnten, analysiert diese Arbeit die meist verbreiteten Methoden, darunter allgemeine Gleichgewichtsmodelle und Asset Pricing Modelle; parametrische, nichtparametrische und semiparametrische Regressionsmodelle; sowie beliebte Black-Box Klassifikationsmethoden. Aufgrund vorteilhafter Eigenschaften binärer Klassifikationsbäume, wie zum Beispiel einer herausragenden Interpretationsmöglichkeit von Entscheidungsregeln, wird der Kern des Handelsalgorithmus unter Verwendung dieser modernen, nichtparametrischen Methode konstruiert. Die optimale Größe des Baumes wird als der entscheidende Faktor für die Vorhersageperformance von Klassifikationsbäumen angesehen. Während eine Vielfalt alternativer populärer Bauminduktions- und Pruningtechniken existiert, die in dieser Studie kritisch gewürdigt werden, besteht eines der Hauptanliegen dieser Arbeit in einer neuartigen Methode asymmetrischen Baumprunings mit Abweisungsoption. Diese Methode wird als Best Node Selection (BNS) bezeichnet. Eine wichtige inverse Fortpflanzungseigenschaft der BNS wird bewiesen. Diese eröffnet eine einfache Möglichkeit, um die Suche der optimalen Baumgröße in der Praxis zu implementieren. Das traditionelle costcomplexity Pruning zeigt eine ähnliche Performance hinsichtlich der Baumgenauigkeit verglichen mit beliebten alternativen Techniken, und es stellt die Standard Pruningmethode für viele Anwendungen dar. Die BNS wird mit cost-complexity Pruning empirisch verglichen, indem zwei rekursive Portfolios aus DAX-Aktien zusammengestellt werden. Vorhersagen über die Performance für jede einzelne Aktie werden von Entscheidungsbäumen gemacht, die aktualisiert werden, sobald neue Marktinformationen erhältlich sind. Es wird gezeigt, dass die BNS der traditionellen Methode deutlich überlegen ist, und zwar sowohl gemäß den Backtesting Ergebnissen als auch nach dem Diebold-Marianto Test für statistische Signifikanz des Performanceunterschieds zwischen zwei Vorhersagemethoden. Ein weiteres neuartiges Charakteristikum dieser Arbeit liegt in der Verwendung individueller Entscheidungsregeln für jede einzelne Aktie im Unterschied zum traditionellen Zusammenfassen lernender Muster. Empirische Daten in Form individueller Entscheidungsregeln für einen zufällig ausgesuchten Zeitpunkt in der Überprüfungsreihe rechtfertigen diese Methode. / Stock picking is the field of financial analysis that is of particular interest for many professional investors and researchers. There is a lot of research evidence supporting the fact that stock returns can effectively be forecasted. While various modeling techniques could be employed for stock price prediction, a critical analysis of popular methods including general equilibrium and asset pricing models; parametric, non- and semiparametric regression models; and popular black box classification approaches is provided. Due to advantageous properties of binary classification trees including excellent level of interpretability of decision rules, the trading algorithm core is built employing this modern nonparametric method. Optimal tree size is believed to be the crucial factor of forecasting performance of classification trees. While there exists a set of widely adopted alternative tree induction and pruning techniques, which are critically examined in the study, one of the main contributions of this work is a novel methodology of nonsymmetrical tree pruning with reject option called Best Node Selection (BNS). An important inverse propagation property of BNS is proven that provides an easy way to implement the search for the optimal tree size in practice. Traditional cost-complexity pruning shows similar performance in terms of tree accuracy when assessed against popular alternative techniques, and it is the default pruning method for many applications. BNS is compared with costcomplexity pruning empirically by composing two recursive portfolios out of DAX30 stocks. Performance forecasts for each of the stocks are provided by constructed decision trees that are updated when new market information becomes available. It is shown that BNS clearly outperforms the traditional approach according to the backtesting results and the Diebold-Mariano test for statistical significance of the performance difference between two forecasting methods. Another novel feature of this work is the use of individual decision rules for each stock as opposed to pooling of learning samples, which is done traditionally. Empirical data in the form of provided individual decision rules for a randomly selected time point in the backtesting set justify this approach.