本研究主要應用基於總體經驗模態分解法(EEMD)之倒傳遞類神經網路(BPNN)預測兩種不同的非線性時間序列數據，包括政大逐時用電量以及逐日歷史黃金價格。透過EEMD，這兩種資料會分別被拆解為數條具有不同物理意義的本徵模態函數(IMF)，而這讓我們可以將這些IMF視為各種影響資料的重要因子，並且可將拆解過後的IMF放入倒傳遞類神經網路中做訓練。 另外在本文中，我們也採用移動視窗法作為預測過程中的策略，另外也應用內插法和外插法於逐時用電量的預測。內插法主要是用於補點以及讓我們的數據變平滑，外插法則可以在某個範圍內準確預測後續的趨勢，此兩種方法皆對提升預測準確度占有重要的影響。 利用本文的方法，可在預測的結果上得到不錯的準確性，但為了進一步提升精確度，我們利用多次預測的結果加總平均，然後和只做一次預測的結果比較，結果發現多次加總平均後的精確度的確大幅提升，這是因為倒傳遞類神經網路訓練過程中其目標為尋找最小誤差函數的關係所致。 / In this paper, we applied the Ensemble Empirical Mode Decomposition (EEMD) based Back-propagation Neural Network (BPNN) learning paradigm to two different topics for forecasting: the hourly electricity consumption in NCCU and the historical daily gold price. The two data series are both non-linear and non-stationary. By applying EEMD, they were decomposed into a finite, small number of meaningful Intrinsic Mode Functions (IMFs). Depending on the physical meaning of IMFs, they can be regarded as important variables which are input into BPNN for training. We also use moving-window method in the prediction process. In addition, cubic spline interpolation as well as extrapolation as our strategy is applied to electricity consumption forecasting, these two methods are used for smoothing the data and finding local trend to improve accuracy of results. The prediction results using our methods and strategy resulted in good accuracy. However, for further accuracy, we used the ensemble average method, and compared the results with the data produced without applying the ensemble average method. By using the ensemble average, the outcome was more precise with a smaller error, it results from the procedure of finding minimum error function in the BPNN training.

在這份研究中，我們提出一個類分散式神經網路框架，此框架為高頻交易系統研究下之子專案。在系統中，我們透過資料探勘程序發掘財務時間序列中的模式，其中所採用的資料探勘演算法之一即為類神經網路。我們實作一個在分散式平台上訓練類神經網路的框架。我們採用Apache Spark來建立底層的運算叢集，因為它提供高效能的記憶體內運算（in-memory computing）。我們分析一些分散式後向傳導演算法（特別是用來預測財務時間序列的），加以調整，並將其用於我們的框架。我們提供了許多細部的選項，讓使用者在進行類神經網路建模時有很高的彈性。 / In this research, we introduce a distributed framework of artificial neural network (ANN) as a subproject under the research of a high-frequency trading (HFT) system. In the system, ANNs are used in the data mining process for identifying patterns in financial time series. We implement a framework for training ANNs on a distributed computing platform. We adopt Apache Spark to build the base computing cluster because it is capable of high performance in-memory computing. We investigate a number of distributed backpropagation algorithms and techniques, especially ones for time series prediction, and incorporate them into our framework with some modifications. With various options for the details, we provide the user with flexibility in neural network modeling.

協同規劃、預測與再補貨(Collaborative Planning, Forecasting and Replenishment; CPFR)，是目前供應鏈管理下重要的討論議題；台灣近年來由於加入WTO與製造業外移使競爭壓力加劇，全球運籌需求提升，使廠商間的合作更加密切，且近年來企業資訊環境與基礎建設逐漸成熟，有助於協同商務之發展。在CPFR流程與供應鏈協同作業環境下，一個供需雙方協同且績效良好的銷售預測具有關鍵的重要性，是管理決策與協同合作時的重要依據；但是多數的企業並沒有一個結構化、有系統化的預測流程及方法，進行多點且不同方法之預測，這樣的銷售預測較無穩定的品質，亦較難提供管理者合理的數據解釋。 在CPFR流程下，強調買賣雙方透過完整、即時資訊的交流，進行短期、單一銷售預測，以提供雙方後續訂單預測、訂單補貨等決策的依據。本研究利用演算法（類神經網路和演化策略法）找出更適合混合性預測架構的解釋變數，再以較適合於實數解之演化策略法於修改黃蘭禎（2004）的三階段之預測模型架構，最後採用實驗方法，進行模型績效驗證。 / Collaborative Planning, forecasting and replenishment (CPFR) is an important issue of supply chain management currently. Because of the severer competition resulted from entrance into WTO and industry integration, cooperation between Taiwanese companies becomes more intensely; enterprises’ information environment and foundation construction attain to maturity also boost the development of collaboration business. In CPRF process and supply chain operation environment, it is critical that a good performance sale forecasting collaborated by both supplier and buyer sides, and it is also the basis of policy decision and collaboration. However, the majority of the companies lack for a structural and systematical forecasting process to proceed with a multi-points forecasting with different methods. This kind of sale forecasting is less of stable quality and is harder to provide the managers a reasonable statistics explanation. Under the CPRF process, both buyers and sellers are able to obtain the short-term and single sale forecasting by real time information communication. Furthermore, the follow-up order forecasting and replenishment strategy decision can be also established through this process. This research finds the variables that are more suitable to the mixed structure by usage of the algorithms, ANN and Evolution Strategy. And this research uses Evolution Strategy that is more suitable to real question to improve the mixed structure of Huang (2004). In the end, experimentation is adopted in order to verify the performance of the model.

增長層級式自我組織映射網路(GHSOM)屬於一種非監督式類神經網路，為自我組織映射網路(SOM)的延伸，擅長於對樣本分群，以輔助分析樣本族群裡的共同特徵，並且可以透過族群間存在的空間關係假設來建立分類器，進而辨別出異常的資料。 因此本研究提出一個創新的對偶方法(即為一個建立決策支援系統架構的方法)分別對舞弊與非舞弊樣本分群，首先兩類別之群組會被配對，即辨識某一特定無弊群體的非舞弊群體對照組，針對這些配對族群，套用基於不同空間假設所設立的分類規則以檢測舞弊與非舞弊群體中是否有存在某種程度的空間關係，此外並對於舞弊樣本的分群結果加入特徵萃取機制。分類績效最好的分類規則會被用來偵測受測樣本是否有舞弊的嫌疑，萃取機制的結果則會用來標示有舞弊嫌疑之受測樣本的舞弊行為特徵以及相關的輸入變數，以做為後續的決策輔助。 更明確地說，本研究分別透過非舞弊樣本與舞弊樣本建立一個非舞弊GHSOM樹以及舞弊GHSOM樹，且針對每一對GHSOM群組建立分類規則，其相應的非舞弊/舞弊為中心規則會適應性地依循決策者的風險偏好最佳化調整規則界線，整體而言較優的規則會被決定為分類規則。非舞弊為中心的規則象徵絕大多數的舞弊樣本傾向分布於非舞弊樣本的周圍，而舞弊為中心的規則象徵絕大多數的非舞弊樣本傾向分布於舞弊樣本的周圍。 此外本研究加入了一個特徵萃取機制來發掘舞弊樣本分群結果中各群組之樣本資料的共同特質，其包含輸入變數的特徵以及舞弊行為模式，這些資訊將能輔助決策者(如資本提供者)評估受測樣本的誠實性，輔助決策者從分析結果裡做出更進一步的分析來達到審慎的信用決策。 本研究將所提出的方法套用至財報舞弊領域(屬於財務舞弊偵測的子領域)進行實證，實驗結果證實樣本之間存在特定的空間關係，且相較於其他方法如SVM、SOM+LDA和GHSOM+LDA皆具有更佳的分類績效。因此顯示本研究所提出的機制可輔助驗證財務相關數據的可靠性。此外，根據SOM的特質，即任何受測樣本歸類到某特定族群時，該族群訓練樣本的舞弊行為特徵將可以代表此受測樣本的特徵推論。這樣的原則可以用來協助判斷受測樣本的可靠性，並可供持續累積成一個舞弊知識庫，做為進一步分析以及制定相關信用決策的參考。本研究所提出之基於對偶方法的決策支援系統架構可以被套用到其他使用財務數據為資料來源的財務舞弊偵測情境中，作為輔助決策的基礎。 / The Growing Hierarchical Self-Organizing Map (GHSOM) is extended from the Self-Organizing Map (SOM). The GHSOM’s unsupervised learning nature such as the adaptive group size as well as the hierarchy structure renders its availability to discover the statistical salient features from the clustered groups, and could be used to set up a classifier for distinguishing abnormal data from regular ones based on spatial relationships between them. Therefore, this study utilizes the advantage of the GHSOM and pioneers a novel dual approach (i.e., a proposal of a DSS architecture) with two GHSOMs, which starts from identifying the counterparts within the clustered groups. Then, the classification rules are formed based on a certain spatial hypothesis, and a feature extraction mechanism is applied to extract features from the fraud clustered groups. The dominant classification rule is adapted to identify suspected samples, and the results of feature extraction mechanism are used to pinpoint their relevant input variables and potential fraud activities for further decision aid. Specifically, for the financial fraud detection (FFD) domain, a non-fraud (fraud) GHSOM tree is constructed via clustering the non-fraud (fraud) samples, and a non-fraud-central (fraud-central) rule is then tuned via inputting all the training samples to determine the optimal discrimination boundary within each leaf node of the non-fraud (fraud) GHSOM tree. The optimization renders an adjustable and effective rule for classifying fraud and non-fraud samples. Following the implementation of the DSS architecture based on the proposed dual approach, the decision makers can objectively set their weightings of type I and type II errors. The classification rule that dominates another is adopted for analyzing samples. The dominance of the non-fraud-central rule leads to an implication that most of fraud samples cluster around the non-fraud counterpart, meanwhile the dominance of fraud-central rule leads to an implication that most of non-fraud samples cluster around the fraud counterpart. Besides, a feature extraction mechanism is developed to uncover the regularity of input variables and fraud categories based on the training samples of each leaf node of a fraud GHSOM tree. The feature extraction mechanism involves extracting the variable features and fraud patterns to explore the characteristics of fraud samples within the same leaf node. Thus can help decision makers such as the capital providers evaluate the integrity of the investigated samples, and facilitate further analysis to reach prudent credit decisions. The experimental results of detecting fraudulent financial reporting (FFR), a sub-field of FFD, confirm the spatial relationship among fraud and non-fraud samples. The outcomes given by the implemented DSS architecture based on the proposed dual approach have better classification performance than the SVM, SOM+LDA, GHSOM+LDA, SOM, BPNN and DT methods, and therefore show its applicability to evaluate the reliability of the financial numbers based decisions. Besides, following the SOM theories, the extracted relevant input variables and the fraud categories from the GHSOM are applicable to all samples classified into the same leaf nodes. This principle makes that the extracted pre-warning signal can be applied to assess the reliability of the investigated samples and to form a knowledge base for further analysis to reach a prudent decision. The DSS architecture based on the proposed dual approach could be applied to other FFD scenarios that rely on financial numbers as a basis for decision making.

本系統係藉由分析歷史交易資料來預測S&P500期貨市場之漲幅。 我們改進了Tsaih et al. (1998)提出的混和式AI系統。 該系統結合了Rule Base 系統以及類神經網路作為其預測之機制。我們針對該系統在以下幾點進行改善：(1) 將原本的日期資料改為使用分鐘資料作為輸入。(2) 本研究採用了“移動視窗”的技術，在移動視窗的概念下，每一個視窗我們希望能夠在60分鐘內訓練完成。(3)在擴增了額外的變數 – VIX價格做為系統的輸入。(4) 由於運算量上升，因此本研究利用TensorFlow 以及GPU運算來改進系統之運作效能。 我們發現VIX變數確實可以改善系統之預測精準度，但訓練的時間雖然平均低於60分鐘，但仍有部分視窗的時間會小幅超過60分鐘。 / The system is made to predict the Futures’ trend through analyzing the transaction data in the past, and gives advices to the investors who are hesitating to make decisions. We improved the system proposed by Tsaih et al. (1998), which was called hybrid AI system. It was combined with rule-based system and artificial neural network system, which can give suggestions depends on the past data. We improved the hybrid system with the following aspects: (1) The index data are changed from daily-based in into the minute-based in this study. (2) The “moving-window” mechanism is adopted in this study. For each window, we hope we can finish training in 60 minutes. (3) There is one extra variable VIX, which is calculated by the VIX in this study. (4) Due to the more computation demand, TensorFlow and GPU computing is applied in our system. We discover that the VIX can obviously has positively influence of the predicting performance of our proposed system. The average training time is lower than 60 minutes, however, some of the windows still cost more than 60 minutes to train.

金融市場是個變化莫測的環境，看似隨機，在隨機中卻隱藏著某些特性與關係。不論是自然現象中的氣象預測或是金融領域中對下一時刻價格的預測, 都有相似的複雜性。 時間序列的預測一直都是許多領域中重要的項目之一, 金融時間序列的預測也不例外。在本論文中我們針對金融時間序列的非線性與非穩態關係引入類神經網路(ANNs) 與集合經驗模態分解法(EEMD), 藉由ANNs處理非線性問題的能力與EEMD處理時間序列信號的優點，並進一步與傳統上使用於金融時間序列分析的自回歸滑動平均模型(ARMA)進行複合式的模型建構，引入燭型圖概念嘗試進行高頻下的台指期貨TAIEX交易。在不計交易成本的績效測試下本研究的高頻交易模型有突出的績效，證明以ANNs、EEMD方法與ARMA組成的混合式模型在高頻時間尺度交易下有相當的發展潛力，具有進一步發展的價值。在處理高頻時間尺度下所產生的大型數據方面，引入平行運算架構SPMD(single program, multiple data)以增進其處理大型資料下的運算效率。本研究亦透過分析高頻時間尺度的本質模態函數(IMFs)探討在高頻尺度下影響台指期貨價格的因素。 / Financial market is complex, unstable and non-linear system, it looks like have some principle but the principle usually have exception. The forecasting of time series always an issue in several field include finance. In this thesis we propose several version of hybrid models, they combine Ensemble Empirical Mode Decomposition (EEMD), Back-Propagation Neural Networks(BPNN) and ARMA model, try to improve the forecast performance of financial time series forecast. We also found the physical means or impact factors of IMFs under high-frequency time-scale. For processing the massive data generated by high-frequency time-scale, we pull in the concept of big data processing, adopt parallel computing method ”single program, multiple data (SPMD)” to construct the model improve the computing performance. As the result of backtesting, we prove the enhanced hybrid models we proposed outperform the standard EEMD-BPNN model and obtain a good performance. It shows adopt ANN, EEMD and ARMA in the hybrid model configure for high-frequency trading modeling is effective and it have the potential of development.