金融市場瞬息萬變，幾乎所有商品價格都是非線性的動態過程，如何預測價格一直都是倍受討論和研究的議題。隨著電腦科技的不斷進步,許多財務學者以市場上的歷史交易資料作為研究對象，希望能夠預測出有效的結果。本研究利用 EEMD 法拆解原始加權指數訊號，建立類神經網路模型，並預測出未來市場之價格後，利用 FK 值當作交易門檻，帶回台指選擇權做交易測試並計算報酬。由於不同神經元個數會配適出不同的預測結果，本研究希望能夠找到較適合使用在指數預測的網路架構。 / The financial market forecasting is characterized by data intensity, noise, non-stationary, high degree of uncertainty, and hidden relationships. Investors are concerned about the forecasting market price. Throughout the development of computational technology, researchers have been involved in data mining on historical trading enabling them to have a more accurate data. This research uses Ensemble Empirical Mode Decomposition-based Artificial Neural Networks (ANNs) learning paradigm to provide different ways to analyze the stock market. In our research, we used the ANN method to obtain our prediction of the stock price. First, the previous day’s stock price needs to be decomposed in order to see the various variables, that is, the numerous IMFs seen on the graphs. Acquiring the information, it is inserted into the ANN method to get a prediction. Following that, the prediction can then be transformed into a simpler result via the Forward Calculator % K indicator. As a result, the FK value can display a signal if to buy or sell, and confirm trading time, and make buy or sell Call-Put decisions on TAIEX options. In summary,we found different neuron numbers in the hidden layers that may affect the result of prediction.

Das Tianshan-Gebirge, als „Wasserturm“ Zentralasiens, hat entscheidenden Einfluss auf die Wasserressourcen der Region. Untersuchungen von 1950 bis 2016 zeigen, dass der Jahresniederschlag in den meisten Teilen des Gebirges zunahm, außer im westlichen Tianshan, wo er abnahm. Es gibt hoch- und niedrigfrequente Schwankungen im Niederschlag mit 3-, 6-, 12- und 27-jährigen Quasiperioden. Auf Dekadenskala gab es zwei Trockenperioden (1950–1962, 1973–1984) und zwei Feuchtperioden (1962–1972, 1985–2016). Seit 2004 ist eine kontinuierliche Feuchtezunahme zu verzeichnen. Zusammenhänge wurden zwischen Zirkulationsmustern und dem Niederschlag identifiziert. Das East Atlantic-West Russia (EATL/WRUS)-Muster korreliert positiv mit dem Winter-Niederschlag. Das Scandinavia (SCAND)-Muster beeinflusst den Sommerniederschlag. Das Silk Road-Muster (SRP) war im Zeitraum 1964-1984 relevant. Die Feuchtigkeitsquellen für den Tianshan-Niederschlag stammen zu 93,2% von kontinentalen Quellen und nur begrenzt aus dem Ozean. Zentralasien ist die Hauptfeuchtequelle für das Gebirge. Im westlichen Tianshan kommt die Feuchtigkeit hauptsächlich von Zentralasien von April bis Oktober und von Westasien von November bis März. Im östlichen Tianshan tragen Ost- und Südasien sowie Sibirien konstant zur Feuchtigkeit im Sommer bei. Der Beitrag der Feuchtigkeit aus dem Nordatlantik zum Sommerniederschlag im nördlichen, zentralen und östlichen Tianshan zeigt einen abnehmenden Trend, obwohl dieser Beitrag ohnehin begrenzt ist. In Monaten mit extremem Winterniederschlag stammt die größte Zunahme der Feuchtigkeit im westlichen Tianshan aus Westasien, während Europa einen wichtigen Beitrag zu den extremen Winterniederschlägen im nördlichen Tianshan leistet. Im östlichen Tianshan ist die Feuchtigkeitszufuhr aus Ost- und Südasien sowie aus Sibirien während der extremen Niederschlagsmonate sowohl im Winter als auch im Sommer erhöht. / The Tianshan Mountains, the "water tower" of Central Asia, are crucial water sources. Precipitation variability and water vapor transport impact water distribution. The study assessed 1950-2016 precipitation using Mann-Kendall tests and EEMD on GPCC data. Multi-timescale precipitation variations were analyzed with NCEP/NCAR reanalysis, and moisture sources during 1979–2017 with ERA–Interim data. Most of Tianshan had increasing annual precipitation, except Western Tianshan, which experienced a downtrend. Precipitation exhibited 3- and 6-year cycles and 12- and 27-year cycles. On the decadal scale, two dry and two wet periods occurred, with continuous humidification since 2004. A significant positive correlation was found between East Atlantic-West Russia EATL/WRUS circulation pattern and winter precipitation. SCAND influenced Tianshan's summer precipitation, with a wet period after 1988 due to enhanced water vapor flux. SCAND and EAP strengthened water vapor fluxes to Tianshan. SRP impacted Tianshan's summer precipitation during 1964–1984. About 93.2% of Tianshan's moisture comes from continental sources. Central Asia dominates moisture supply. Western Tianshan receives moisture mainly from Central Asia (April to October) and Western Asia (November to March). Almost 13.0% of Eastern Tianshan's summer moisture originates from East and South Asia and Siberia, with steady contributions. Moisture from the North Atlantic Ocean to summer precipitation in Northern, Central, and Eastern Tianshan shows a decreasing trend, but limited overall contribution. Extreme winter precipitation in Western Tianshan is linked to moisture from West Asia. Europe significantly contributes to extreme winter precipitation in Northern Tianshan. Eastern Tianshan sees enhanced moisture from East and South Asia and Siberia during extreme precipitation months in winter and summer.

The aim of this thesis is to introduce with principle of Empirical Mode Decomposition method and possibility use for correction of baseline wander in ECG signals. The thesis describes the main components of the ECG signal, a selection of possible types of signal noise, its property and principles of chosen methods for filtration of ECG signals. In conclusion the evaluation of the effectiveness of the EMD method for filtering a baseline wander and it comparing with effectiveness of the linear filtration. Functionality of used algorithms has been tested on signals of CSE standard library.

金融市場常常受到政治、經濟與社會環境等因素所影響，所得到價格為眾多變數交互作用的結果，包含了許多雜訊。本文引進一套數據處理方法「整體經驗模態分解」(Ensemble Empirical Mode Decomposition，EEMD)來分析「期貨市場」以及「預測市場」。第一個實證利用EEMD處理台股期貨，分析對台股指數的解釋能力，並同時與原始台股期貨預測台股指數，比較預測結果；第二個實證利用EEMD來分析預測市場，判別是否能有效的消除雜訊，準確預測選舉結果。 第一個實證結果發現，EEMD能有效地過濾期貨市場的雜訊，另外，在最後到期日前十二天或者是前九天，以週期為6.5日經EEMD處理的台股期貨對台股指數的預測較原始台股期貨預測準確；第二個實證結果指出，直接利用EEMD處理預測市場得到的長期趨勢「剩餘訊號」(Residue)來預測選舉並無優於原始預測市場，主因為預測市場參與者不只在乎長期趨勢，亦在乎短期事件的衝擊，故直接利用剩餘訊號預測選舉結果會有所失真，而將剩餘訊號由低頻率之「本質模態函數」(Intrinsic Modes Function，IMF)合併至週期為6日與12日的IMF，得到了EEMD週趨勢價格，分成選前一天和選前十天的資料並與原始預測市場以及民調預測做比較，從不同的準則來看，發現以EEMD週趨勢價格來做選舉預測，準確度較原始預測市場與民調預測的結果更好。根據中選會2012年初選前對選罷法做成的解釋，未來事件交易所在選前十日亦須停止交易，我們可將EEMD運用在日後的選舉預測，把預測市場的合約價格以EEMD處理，應可提高選舉預測的準確度。 / The financial markets are usually affected by political, economic and social environment factors, and thus the volatilities of asset prices in these markets are subject to a lot of noises and shocks. To filter out noises and quantify shocks, this paper applies a data processing method, Ensemble Empirical Mode Decomposition (EEMD), and demonstrates its improved prediction to the futures and election prediction markets. While the first empirical application shows that the EEMD effectively filters out the noises in the futures market, the second one indicates that the Taiwanese election prediction using EEMD “residue” is not as accurate as that by original data from the prediction market. The reason why the residue cannot serve as a good predictor is that the market participants consider not only the long-term trend, but also shocks, especially those right before the elections. We then attempt to predict the election outcomes by the week trend series processed by EEMD. The prediction by the week EEMD trend series turns out to be more accurate than that by the poll and original prediction market. Based on this study, we can apply the EEMD to the next election prediction and improve its accuracy.