Browsing by Author "Chen, An-Pin"

Now showing 1 - 20 of 43

A clustering time series model for the optimal hedge ratio decision making
(2014-08-22)
Hsu, Yu-Chia
;
Chen, An-Pin
;
資訊管理與財務金融系註：原資管所+財金所
;
Department of Information Management and Finance
In this study, a novel procedure combining computational intelligence and statistical methodologies is proposed to improve the accuracy of minimum-variance optimal hedge ratio (OHR) estimation over various hedging horizons. The time series of financial asset returns are clustered hierarchically using a growing hierarchical self-organizing map (GHSOM) based on the dynamic behaviors of market fluctuation extracted by measurement of variances, covariance, price spread, and their first and second differences. Instead of using original observations, observations with similar patterns in the same cluster and weighted by a resample process are collected to estimate the OHR. Four stock market indexes and related futures contracts, including Taiwan Weighted Index (TWI), Standard & Poor\'s 500 Index (S&P 500), Financial Times Stock Exchange 100 Index (FUSE 100), and NIKKEI 255 Index, are adopted in empirical experiments to investigate the correlation between hedging horizon and performance. Results of the experiments demonstrate that the proposed approach can significantly improve OHR decisions for mid-term and long-term hedging compared with traditional ordinary least squares and naive models. (C) 2014 Elsevier B.V. All rights reserved.
A collaboration environment for R&D project
(2007)
Liao, August
;
Fu, Li-Dien
;
Chen, An-Pin
;
資訊管理與財務金融系註：原資管所+財金所
;
Department of Information Management and Finance
The research question here is "how to effectively integrate the R&D experts and software developers to realize an effective R&D project collaboration software?" This research conducted a case study on the deployment of an integrated development methodology in a world-class semiconductor manufacture company. From the case study, the integrated methodology, Development Collaboration Diamond Model (DCDM), was designed, implemented and obtained the dramatic performance.
A Hybrid Forecasting Model for Foreign Exchange Rate Based on a Multi-neural Network
(2008)
Chen, An-Pin
;
Hsu, Yu-Chia
;
Hu, Ko-Fei
;
資訊管理與財務金融系註：原資管所+財金所
;
Department of Information Management and Finance
In this work, a multi-neural network model consisting of three sub-networks and one master network is proposed to combine the fundamental theorem and technical analysis in TWD/USD exchange rate forecasting. The long-term, mid-term, and short-term tendencies of exchange rate are forecasted separately by different sub-networks. Five macro economics factors of price level, interest rates, money supply, imports/exports, and productivity, and seven practical technical indicators of fifteen-day and one-day intervals are selected as the input variables of the three sub-networks. The master network then provides the integrated forecasting according to the three sub-networks. To increase forecasting accuracy, a threshold filtering mechanism was applied in this work. The experiment result shows that the multi-neural network is more effective than the random walk model and single-neural network model, and with the threshold filtering can achieve high accuracy.
A Market Making Quotation Strategy Based on Dual Deep Learning Agents for Option Pricing and Bid-Ask Spread Estimation
(2018-01-01)
Hsu, Pei-Ying
;
Chou, Chin
;
Huang, Szu-Hao
;
Chen, An-Pin
;
資訊管理與財務金融系註：原資管所+財金所
Traditional professional traders and institutional investors utilized complex statistical models to price various derivative contracts and make trading decisions in the option and future markets. In recent years, with the rapid growth of algorithmic trading and program trading, the advanced information and communication technology has become an indispensable element for high-frequency traders, especially for the market makers. In addition, artificial intelligence and deep learning also plays an important role in novel financial technology (FinTech) research field. In this paper, we proposed a market making quotation strategy based on deep learning structure and practical finance domain knowledge. The proposed dual agents will simultaneously model the option prices and bid-ask spreads. The experiments demonstrate that our system can precisely estimate the value of options than famous financial engineering models. It also can be extended to develop proper market making quotation strategies to trade the options of Taiwan Stock Exchange Capitalization Weighted Stock Index(TAIEX).
A new efficient approach for data clustering in electronic library using ant colony clustering algorithm
(2006)
Chen, An-Pin
;
Chen, Chia- Chen
;
資訊管理與財務金融系註：原資管所+財金所
;
Department of Information Management and Finance
Purpose - Traditional library catalogs have become inefficient and inconvenient in assisting library users. Readers may spend much time in searching library materials via printed catalogs. Readers need an intelligent and innovative solution to overcome this problem. The purpose of this paper is to illustrate how data mining technology is a good approach to fulfill readers' requirements. Design/methodology/approach - Data mining is considered to be the nontrivial extraction of implicit, previously unknown, and potentially useful information from data. This paper analyzes the readers' borrowing records by using the following techniques: data analysis, building data warehouse and data mining. Findings - The mining results show that all readers can be categorized into five clusters, and each cluster has its own characteristics. It was also found that the frequency for graduates and associate researchers to borrow multimedia data is much higher. This phenomenon shows that these readers have a higher preference for accepting digitized publication. Besides, we notice that more readers borrow multimedia data rise in years. This up trend indicates that readers are gradually shifting their preference in reading digital publications. Originality/value - The paper proposes a technique to discover clusters by using ant colony methods.
A Novel Modified Particle Swarm Optimization for Forecasting Financial Time Series
(2009)
Chen, An-Pin
;
Huang, Chien-Hsun
;
Hsu, Yu-Chia
;
資訊管理與財務金融系註：原資管所+財金所
;
Department of Information Management and Finance
Time series has been widely applied in the real world, traditional methods can hardly solve the dynamic environment issue resulting from the assumption of stationary process Many traditional models and artificial intelligence technologies had been developed under this assumption, and adapted the dynamic environment based on the time-varying characteristic But these models still has drawback of dividing the time series Into training set and testing set when developing the models It means the time-varying characteristic of these two sets did not be considered, and it might cause spurious regression phenomenon and result in misleading the statistic analysis In order to forecast dynamic time series, a model which can consider the dynamic environment and conquer the out-of-sample problem is necessary Particle swarm optimization (PSO) has the characteristics of fast-convergence and avoiding local optimal, also has been widely used in the time series forecasting In this research, we proposed a modified PSO to consider the dynamic environment issue and use the advantage of PSO to forecast the dynamic financial time series
A RFID-Enabled Traceability System for the Supply Chain of Live Fish
(2008)
Hsu, Yu-Chia
;
Chen, An-Pin
;
Wang, Chun-Hung
;
資訊管理與財務金融系註：原資管所+財金所
;
Department of Information Management and Finance
Traceability is important in the food supply chain to ensure the consumers' food safety, especially for the fresh products. In recent years, many solutions which applied various emerging technology have been proposed to improve the traceability of fresh product. However, the traceability system needs to be customized to satisfy different requirements. The system depends on the different product properties and supply chain models. This paper proposed a RFID-enabled traceability system for live fish supply chain. The system architecture is designed according to the specific requirement gathered in the life fish processing. Likewise, it is adaptive for the small and medium enterprises. The RFID tag is put on each live fish and is regarded as the mediator which links the live fish logistic center, retail restaurants and consumers for identification. The sensors controlled by the PLC are used to collect the information in farming as well as the automatic transporting processes. The traceability information is designed to be exchanged and used on a web-based system for farmers and consumers. The system was implemented and deployed in the live fish logistic center for trial, and the results are valuable for practical reference.
An Integrated Home Financial Investment Learning Environment Applying Cloud Computing in Social Network Analysis
(2011)
Wen, Mao-Ping
;
Lin, Hsio-Yi
;
Chen, An-Pin
;
Yang, Chyan
;
資訊管理與財務金融系註：原資管所+財金所
This paper tried to apply cloud computing technology in social network analysis for a comprehensive home financial learning environment that individual investors may use as a reference in establishing web-based learning and investment platforms. The major contributions were described in three parts. First, this paper advanced the social network analysis technology to be able to handle millions of nodes and links. Second, we demonstrate how cloud computing can be applied to advanced computing in social network. Third, we performed several intelligent analyses on a very popular social network, IHFILE, to identify some interesting and important features of it. In addition to analyzing a homogeneous social network such as IHFILE, we also propose direction of how cloud computing can be performed on a social network analysis as our future work.
An inter-market arbitrage trading system based on extended classifier systems
(2011-04-01)
Hsu, Yu-Chia
;
Chen, An-Pin
;
Chang, Jia-Haur
;
資訊管理與財務金融系註：原資管所+財金所
;
Department of Information Management and Finance
Traditionally, the most popular arbitrage strategy is derived from the cost of carry model or by using the econometrics approach. However, these approaches have difficulty in dealing with intra-day 1-min trading data and capturing inter-market arbitrage opportunity in the real world. In this research, we propose computational intelligence approaches based on the extended classifier system (XCS). First, in order to reduce the amount of data, the original data streams of intra-day 1-min trading data are filtered by the conditions of variant price spread relation. XCS is then adopted for knowledge rule discovery. After analyzing the property with domain-specific knowledge that the price of index futures will get close to that of spot products at the time the futures mature, four important factors related to bias, price spread, expiry date, and intraday trading timing are considered as the conditions of XCS to build the inter-market arbitrage model. The inter-market spread of the Taiwan Stock Index Futures (TX) traded at the Taiwan Futures Exchange (TAIFEX) and the Morgan Stanley Capital International (MSCI) Taiwan Index Futures traded at the Singapore Exchange Limited (SGX) are chosen for an empirical study to verify the accuracy and profitability of the model. (C) 2010 Elsevier Ltd. All rights reserved.
Application of Dynamic Financial Time-Series Prediction on the Interval Artificial Neural Network Approach with Value-at-Risk Model
(2008)
Lin, Hsio-Yi
;
Chen, An-Pin
;
資訊管理與財務金融系註：原資管所+財金所
;
Department of Information Management and Finance
Artificial Neural Networks (ANNs) are promising approaches for financial time-series prediction. This study adopts a hybrid approach, called a Fuzzy BPN, consisting of a Back-Propagation Neural Network (BPN) and a fuzzy membership function which takes advantage of the ANNs\' nonlinear features and interval values instead of the shortcoming of ANNs\' single-point estimation. To employ the two characteristics mentioned above, a dynamic intelligent time-series forecasting system will be built more efficiently for practical financial predictions. Additionally, with the liberalization and opening of financial markets, the relationships among financial commodities became much closer and complicated. Hence, establishing a perfect measure approach to evaluate investment risk has become a critical issue. The objective of this study is not only to achieve higher efficiency in dynamic financial time-series predictions but also a more effective financial risk control with Value-at-Risk methodology, which is called Fuzzy-VaR BPN model in this study. By extending to the financial market environment, it is expected that wider and more suitable applications in financial time-series and risk management problems would be covered. Moreover, the Fuzzy-VaR BPN model would be applied to the Taiwan Top50 Tracker Fund to demonstrate the capability of our study.
Application of Self-Organizing Mapping Neural Network for Discovery of Market Behavior of Equity Fund
(2009)
Chen, Jen-Hua
;
Huang, Chiung-Fen
;
Chen, An-Pin
;
資訊管理與財務金融系註：原資管所+財金所
;
Department of Information Management and Finance
Maximizing the profit and minimizing the loss notwithstanding the trend of the market is always desirable in any investment strategy. The present research develops an investment strategy, which has been verified effective in the real world, by employing self-organizing map neural network for mutual funds and tracking the trends of stock market indices according to macro-economy indicators, weighted indices, and rankings of mutual funds. Our experiment shows if utilizing strategy 3 according to our model during a period from January 2002 to December 2008 the total returns could be at 122 percents even though the weighted index falls 22 percents and averaged investment returns for random transaction strategies stand at minus 25 percents during the same period. As such, we conclude that our model does efficiently increase the investment return.
Applying Market Profile Theory to Analyze Financial Big Data and Discover Financial Market Trading Behavior - A Case Study of Taiwan Futures Market
(2016-01-01)
Huang, Wei-Yuan
;
Chen, An-Pin
;
Hsu, Yu-Hsiang
;
Chang, Hua-Yang
;
Tsai, Ming-Wu
With financial market constantly changing, prices are often affected by many factors that we cannot predict its direction easily especially in the market correction. If investors want to make profits, they must find a relatively low-risk entry points. This study is based on Market Profile Theory to use the displacement of point of control (POC) in different trading days to find out the best extremely short-term entry and exit points in financial big data in order to have experiment and analysis. We expect to find knowledge and behavior of the potential market that can help traders to make profits in extremely short-term trading. And at the end of this study, we can refute that Taiwan Index Futures Market conform to the weak form of efficient market hypothesis. This study found that the POC of historical trading day can be the reference and recommendation of entry point. The greatest performance of making profit is using 5-days historical POC to join the experiment. And it shows POC has the characteristic that the most traders accept its price.
Applying market profile theory to forecast Taiwan Index Futures market
(2014-08-01)
Chen, Chiu-Chin
;
Kuo, Yi-Chun
;
Huang, Chien-Hua
;
Chen, An-Pin
;
資訊管理與財務金融系註：原資管所+財金所
This research applies a market profile to establish an indicator to classify the correlation between the variation in price and value with the stock trends. The indicator and technical index are neural network architecture parameters that assist to extrapolate the market logic and knowledge rules that influence the TAIEX futures market structure via an integral assessment of physical quantities. To implement the theory of market profile on neural network architecture, this study proposes qualitative and quantitative methods to compute a market profile indicator. In addition, the indicator considers the variation and relevance between long-term and short-term trends by incorporating the long-term and short-term change in market in its calculation. An assessment of forecasting performance on different calculation approaches of market profile indicator and technical analysis is conducted to differentiate their accuracies and profitability. The experimental results show the qualitative market profile indicator outperforms the quantitative approach in a short-term forecast period. In contrast, the quantitative market profile indicator has a better trend-predicting ability, thus it is more effective in the long-term forecast period. The integration of market profile and technical analysis surpasses technical analysis as a neural network architecture parameter by effectively improving forecasting performance and profitability. (C) 2014 Elsevier Ltd. All rights reserved.
Applying the Extended Classifier System to Trade Interest Rate Futures Based on Technical Analysis
(2008)
Chen, An-Pin
;
Hsu, Yu-Chia
;
Huang, Chien-Hua
;
Yang, Ya-Chun
;
資訊管理與財務金融系註：原資管所+財金所
In practice, it is difficult to gain profit in the process of trading interest rate derivative commodities. This could be attributed to the complexity of existing pricing models, which are derived from the term structure and yield curve, both of which cannot adapt well to short-term market dynamics. In this study, we use the Extended Classifier System (XCS) to model the market behavior of interest rate futures, the purpose of which is to provide effective trading decision support. Several technical indicators and their first- and second-order derivatives are selected as the market descriptive variables, which are then used for XCS training. Finally, the adaptive rules of the classifiers, which consist of conditions with relative actions considered helpful for constructing the automatic trading system, are generated from the XCS knowledge discovery process. The market data of the 10-year government bond futures traded in Taiwan are chosen for empirical study to verify the accuracy and profitability of the XCS model. These were also used to conduct a comparative evaluation between the random walk and tendency following models and the XCS model.
Binary Classification and Data Analysis for Modeling Calendar Anomalies in Financial Markets
(2016-01-01)
Tung, Hui-Hsuan
;
Chen, Yu-Ying
;
Huang, Szu-Hao
;
Cheng, Chiao-Chun
;
Chen, Yu-Fu
This paper studies on the Day-of-the-week effect by means of several binary classification algorithms in order to achieve the most effective and efficient decision trading support system. This approach utilizes the intelligent data-driven model to predict the influence of calendar anomalies and develop profitable investment strategy. Advanced technology, such as time-series feature extraction, machine learning, and binary classification, are used to improve the system performance and make the evaluation of trading simulation trustworthy. Through experimenting on the component stocks of S&P 500, the results show that the accuracy can achieve 70% when adopting two discriminant feature representation methods, including "multi-day technical indicators" and "intra-day trading profile." The binary classification method based on LDA-Linear Prior kernel outperforms than other learning techniques and provides the investor a stable and profitable portfolios with low risk. In addition, we believe this paper is a FinTech example which combines advanced interdisciplinary researches, including financial anomalies and big data analysis technology.
Building a Learning Games Network in Cloud Learning Platform based on Immigrant Education
(2011)
Tu, Chien-Chih
;
Chen, An-Pin
;
資訊管理與財務金融系註：原資管所+財金所
;
Department of Information Management and Finance
As with rapid growth of the computer technology, E-learning systems usually require many hardware and software resources, There are numerous educational institutions that cannot offer such investments, and cloud learning platform is the best solution for them. This paper proposes a model of using cloud computing and learning network upon cloud-learning solutions development. The cloud-learning platform combined with different types of games is gradually noticed by people because it can enhance user learning motivation. This research has developed a Chinese language cloud-learning system for the new immigrant based on games mode, and investigated the properties of game-based cloud-learning system, expecting to help more and more new immigrants in Taiwan. The basic concept for designing the system is developing the learning games network using digital materials, which is applied to the cloud learning platform to attract the immigrant residents and assist them to improve Chinese language skills.
Cloning Strategies from Trading Records using Agent-based Reinforcement Learning Algorithm
(2018-01-01)
Chen, Chiao-Ting
;
Chen, An-Pin
;
Huang, Szu-Hao
;
資訊管理與財務金融系註：原資管所+財金所
;
Department of Information Management and Finance
Investment decision making is considered as a series of complicated processes, which are difficult to be analyzed and imitated. Given large amounts of trading records with rich expert knowledge in financial domain, extracting its original decision logics and cloning the trading strategies are also quite challenging. In this paper, an agent-based reinforcement learning (RL) system is proposed to mimic professional trading strategies. The concept of continuous Markov decision process (MDP) in RL is similar to the trading decision making in financial time series data. With the specific-designed RL components, including states, actions, and rewards for financial applications, policy gradient method can successfully imitate the expert's strategies. In order to improve the convergence of RL agent in such highly dynamic environment, a pre-trained model based on supervised learning is transferred to the deep policy networks. The experimental results show that the proposed system can reproduce around eighty percent trading decisions both in training and testing stages. With the discussion of the tradeoff between explorations and model updating, this paper tried to fine-tuning the system parameters to get reasonable results. Finally, an advanced strategy is proposed to dynamically adjust the number of explorations in each episode to achieve better results.
Cloud Computing Architecture for Social Computing - A Comparison Study of Facebook and Google
(2011)
Yang, Bo-Wen
;
Tsai, Wen-Chih
;
Chen, An-Pin
;
Ramandeep, Singh
;
資訊管理與財務金融系註：原資管所+財金所
As far as we know, cloud computing is a new business model and the cloud computing architecture is the famous topic recently. Today, everyone enjoy the innovative search engine or social network application for new Internet services no longer require the large capital outlays in hardware to access those service or the human expense to operate it. Currently, Google is the largest search engine and Facebook is the largest social network in the Software as a Service (SaaS). But how them can support the huge requests from world thought each personal computer, mobile device, and smart phone. In this paper, we will try to analysis their backend cloud computing architecture to support future SaaS especially in large social network.
Clustering Time Series Data by SOM for the Optimal Hedge Ratio Estimation
(2008)
Hsu, Yu-Chia
;
Chen, An-Pin
;
資訊管理與財務金融系註：原資管所+財金所
;
Department of Information Management and Finance
The fat-tailed and leptokurtic properties observed in most financial asset return series would cause the inaccuracy of hedge ratio estimation because most traditional statistics approaches are based on the assumption of normal distribution. In this study, a novel approach is proposed using self-organizing map (SOM, also called Kohonen's Self-Organizing Feature Map) for time series data clustering and similar pattern recognition to improve the optimal hedge ratio (OHR) estimation. Five SOM-based models (considering the weight for averaging and the interval for data sampling) and two traditional models (ordinary least square method and naive hedge) were compared in Taiwan stock market hedging. The experiment demonstrates the feasibility of applying SOM, and the empirical results show that SOM approach provides a useful alternative to the OHR estimation.
Constructing a dynamic stock portfolio decision-making assistance model: using the Taiwan 50 Index constituents as an example
(2007-10-01)
Chen, Mei-Chih
;
Lin, Chang-Li
;
Chen, An-Pin
;
資訊管理與財務金融系註：原資管所+財金所
;
Department of Information Management and Finance
There are several decisions in investment management process. Security selection is the most time-consurning stage. Tatical allocation is in order to take advantage of market opportunities based on short-term prediction (Amenc and Le Sourd in Portfolio theory and performance analysis. Wiley, 2003). Although it is difficult to keep track of the fluctuations of volatile financial markets, the capacity of artificial intelligence to perform spatial search and obtain feasible solutions has led to its recent widespread adoption in the resolution of financial problems. Classifier systems possess a dynamic learning mechanism, they can be used to constantly explore environmental conditions, and immediately provide appropriate decisions via self-aware learning. This study consequently employs a classifier system in conjunction with real number encoding to investigate how to obtain optimal stock portfolio based on investor adjustment cycle. We examine the constituents of the TSEC Taiwan 50 Index taking moving average (MA), stochastic indicators (KD), moving averaae convergence divergence (MACD), relative strength index (RSI) and Williams %R (WMS %R) as input factors, adopting investor-determined adjustment cycle to allocate capital, and then constructing stock portfolio. We have conducted empirical testing using weekly and monthly adjustment cycle; the results revealed that this study's decision-making assistance model yields average annual interest rate of 49.35%, which is significantly better than the -6.59% of a random purchase model. This research indicates that a classifier system can effectively monitor market fluctuations and help investors obtain relatively optimal returns. The assistance model proposed in this study thus can provide really helpful decision-making information to investors.