
Tomoya SUZUKIProfessor
■Researcher basic information
Organization
- College of Engineering Department of Mechanical Systems Engineering
- Graduate School of Science and Engineering(Master's Program) Major in Computer and Information Sciences
- Graduate School of Science and Engineerin(Doctoral Program) Major in Society's Infrastructure Systems Science
- School of Collaborative Regional Innovation
- Faculty of Applied Science and Engineering Domain of Mechanical Systems Engineering
Research Areas
Educational Background
Career
- Oct. 2017 - Present, 大和証券投資信託委託(株), クウォンツ運用部, 特任主席研究員
- Apr. 2017 - Present, Ibaraki University, Graduate School of Science and Engineering Department of Science (Chemistry Course), 教授
- Aug. 2018 - Jul. 2024, CollabWiz, Inc., CEO
- Apr. 2016 - Mar. 2017, 茨城大学, 工学部 知能システム工学科, 教授
- Apr. 2009 - Mar. 2016, 茨城大学, 工学部 知能システム工学科, 准教授
- Apr. 2006 - Mar. 2009, 同志社大学, 理工学部 情報システムデザイン学科, 専任講師
- Apr. 2005 - Mar. 2006, 東京電機大学, 工学部 電子工学科, 助手
Member History
External link
Message from Researchers
(Message from Researchers)
Born in Niigata City, Niigata Prefecture. In 2005, he completed a doctoral course at the Department of Physics, Graduate School of Science, Tokyo University of Science. Doctor of Science. In the same year, he became an assistant professor at the Department of Electronic Engineering, Faculty of Engineering, Tokyo Denki University. Since 2006, he has been a full-time lecturer at the Department of Information System Design, Faculty of Engineering, Doshisha University. Since 2017, he has been a specially appointed chief researcher at Daiwa Asset Management Co., Ltd. Quants Investment Department, and since 2018, he has concurrently served as CEO of CollabWiz Co., Ltd. He works on returning research results to society. Full member of Sigma Xi.
■Research activity information
Award
Paper
- Multi-objective optimization support tool for global equity management
Eisuke Sawahata; Yoshitsugu Hanawa; Hiromitsu Kawamata; Tomoya Suzuki
Proc. of RISP International Workshop on NCSP, Feb. 2024, [Reviewed] - Machine learning of economic sensitive industries for domestic equity management
Nozomu Orita; Takashi Suzuki; Tomoya Suzuki
Proc. of RISP International Workshop on NCSP, Feb. 2024, [Reviewed] - Visualization of Searching System of Thematic Equity Funds
Ziyi Dai; Tomoya Suzuki
Proc. of RISP International Workshop on NCSP, Feb. 2024, [Reviewed] - Long-Term Modeling of Financial Machine Learning for Active Portfolio Management
Kazuki Amagai; Tomoya Suzuki
Journal of Signal Processing, Oct. 2023, [Reviewed] - Composition of Thematic Equity Funds by Searching Multi and Unknown Words
Zijie Luo; Ziyi Dai; Wataru Kuramoto; Tomoya Suzuki
Proc. of RISP International Workshop on NCSP, 01 Mar. 2023, [Reviewed] - Long-Term Modeling of Financial Machine Learning with Multiple Time Scales
Kazuki Amagai; Riku Tanaka; Tomoya Suzuki
Proc. of RISP International Workshop on NCSP, 01 Mar. 2023, [Reviewed] - Forex Trading Strategy That Might Be Executed Due to the Popularity of Gotobi Anomaly
Hiroki Bessho; Takanari Sugimoto; Tomoya Suzuki
Proc. of RISP International Workshop on NCSP, 01 Mar. 2023, [Reviewed] - Forex Trading Strategy That Might Be Executed Due to the Popularity of Gotobi Anomaly
Hiroki Bessho; Takanari Sugimoto; Tomoya Suzuki
arXiv, 01 Feb. 2023 - 機械学習による為替フォワード取引期間の判別モデルおよび運用シミュレーション
雉子波晶; 杉本誠忠; 酒本隆太; 鈴木智也, Last
JAFEEジャーナル, 01 Jun. 2022, [Reviewed] - Nonlinear modeling for equity valuation by machine learning
Yuuki Tsukahara; Riku Tanaka; Tomoya Suzuki, Last
Nonlinear Theory and its Applications, IEICE, 01 Apr. 2022, [Reviewed] - Visualization of nonlinear relationship in capital flows of Japanese mutual funds
Takuma Nakamichi; Ryouhei Yoshida; Riku Tanaka; Tomoya Suzuki, Last
Nonlinear Theory and its Applications, IEICE, 01 Apr. 2022, [Reviewed] - Visualization of Nonlinear Relationship Hidden in Money Flows of Japanese Mutual Funds
Takuma Nakamichi; Ryouhei Yoshida; Riku Tanaka; Tomoya Suzuki, Last
Proc. of Nonlinear Science Workshop, 08 Oct. 2021, [Reviewed] - Nonlinear Modeling for Equity Valuation by Machine Learning
Yuuki Tsukahara; Riku Tanaka; Tomoya Suzuki, Last
Proc. of Nonlinear Science Workshop, 08 Oct. 2021, [Reviewed] - 機械学習による中古車落札価格の要因分析及び割安評価
工藤大輝; 福西亮介; 黛広樹; 鈴木智也, Last
情報処理学会論文誌, 01 Jul. 2021, [Reviewed] - 国内輸入に伴う貿易取引通貨比率とゴトオビアノマリーの関係
秋山朋也; 杉本誠忠; 酒本隆太; 鈴木智也, Last
JAFEEジャーナル, 01 Jul. 2021, [Reviewed] - 暗号資産の異常ジャンプ検知による分散投資
鈴木智也; 玉城玲奈, Lead
Technical Analysts Journal, Jul. 2020, [Reviewed] - カバー先銀行の集合知による外国為替レートの短期予測
信号処理学会論文誌, 01 May 2020, [Reviewed] - Direct Prediction of individual Contract Prices at Auto Auction with Deep Neural Network
Yuuma Hayami; Hiromichi Sakurai; Daiki Kudou; Eriko Hasegawa; Rikizou Shimoyama; Ryosuke Fukunishi; Hiroki Mayuzumi; Tomoya Suzuki, Corresponding
Proceedings of International Symposium on Nonlinear Circuits and Signal Processing, Mar. 2019, [Reviewed] - Prediction of Contract Prices at Auto Auction with Time Series Models
Risa Yamashita; Hiromichi Sakurai; Yuma Hayami; Eriko Hasegawa; Rikizou Shimoyama; Ryosuke Fukunishi; Hiroki Mayuzumi; Tomoya Suzuki, Corresponding
Proceedings of International Symposium on Nonlinear Circuits and Signal Processing, Mar. 2019, [Reviewed] - Prediction of Foreign Exchange Best Rates by Using Collective Knowledge of Counterparty Banks
Kazuto Yano; Takehiro Suzuki; Tomoya Suzuki, Corresponding
Proceedings of International Symposium on Nonlinear Circuits and Signal Processing, Mar. 2019, [Reviewed] - Auto-extraction of Influential Keywords Included in Financial News Headlines
Masahiro Miyoshi; Wenkai Shi; Yui Hosoki; Junichi Eguchi; Minoru Sasaki; Tomoya Suzuki, Corresponding
Proceedings of International Symposium on Nonlinear Circuits and Signal Processing, Mar. 2019, [Reviewed] - 非線形ポートフォリオモデルにおける主成分分析の活用
柳澤和輝; 鈴木智也, Corresponding
電子情報通信学会論文誌, 01 May 2018, [Reviewed] - Consensus Ratio and Two-steps Selection to Detect Profitable Stocks: Modern Technical Analysis Using Machine Learning Approach
Tomoya Suzuki, Lead
International Federation of Technical Analysts (IFTA) Journal, 2018, [Reviewed] - Predictability of Financial Market Indexes by Deep Neural Network
Tomoya Onizawa; Takehiro Suzuki; Tomoya Suzuki, Corresponding
Proceedings of International Symposium on Nonlinear Theory and its Applications, Dec. 2017, [Reviewed] - Technical Trading Strategy Using Reactions to Stock Price Jumps
Tokimaru Tsuruta; Tomoya Suzuki, Corresponding
Proceedings of International Symposium on Nonlinear Theory and its Applications, 30 Nov. 2016, [Reviewed] - Principal Component Stock Portfolio Based on Nonlinear Prediction
Kazuki Yanagisawa; Tomoya Suzuki, Corresponding
Proceedings of International Symposium on Nonlinear Theory and its Applications, 30 Nov. 2016, [Reviewed] - Biased Reactions to Abnormal Stock Prices Detected by Autoencode
Hiroyuki Gotou; Tomoya Suzuki, Corresponding, To detect abnormal price jumps of financial markets, some indicators based on volatility have been used such as the bipower variation and the BPV ratio. However, these indicators only focus on a single individual stock and do not consider the relationships among all individual stocks composing a complex financial system. For this reason, we applied an autoencoder to learn the relationships among all stocks, and we considered a stock price that the autoencoder cannot restore as an abnormal price. Moreover, we identified that the price movement immediately following an abnormal price is clearly biased, and we confirmed the validity of our trading strategy based on this anomaly by performing some statistical significance tests., Research Institute of Signal Processing, Japan
Proceedings of International Symposium on Nonlinear Theory and its Applications, 30 Nov. 2016, [Reviewed] - Detection of Abnormal Stock Prices with Autoencoder
Hiroyuki Gotou; Tomoya Suzuki
Proceedings of International Symposium on Nonlinear Circuits and Signal Processing, Mar. 2016, [Reviewed] - Technical Trading Strategy Using the Reaction to Price Jumps in American Stock Market
Tokimaru Tsuruta; Hiroya Koizumi; Tomoya Suzuki
Proceedings of International Symposium on Nonlinear Circuits and Signal Processing, Mar. 2016, [Reviewed] - Evidence of Enhancing Nonlinear Predictability of Stock Price Movements by the Principal Component Analysis
Kazuki Yanagisawa; Tomoya Suzuki, Corresponding
Proceedings of International Symposium on Nonlinear Circuits and Signal Processing, Mar. 2016, [Reviewed] - Biased Reactions to Abnormal Stock Prices Detected by Autoencoder,'' Journal of Signal Processing
Hiroyuki Gotou; Tomoya Suzuki, Corresponding, To detect abnormal price jumps of financial markets, some indicators based on volatility have been used such as the bipower variation and the BPV ratio. However, these indicators only focus on a single individual stock and do not consider the relationships among all individual stocks composing a complex financial system. For this reason, we applied an autoencoder to learn the relationships among all stocks, and we considered a stock price that the autoencoder cannot restore as an abnormal price. Moreover, we identified that the price movement immediately following an abnormal price is clearly biased, and we confirmed the validity of our trading strategy based on this anomaly by performing some statistical significance tests., Research Institute of Signal Processing, Japan
Journal of Signal Processing, 2016, [Reviewed] - 金融市場のジャンプに対する反応を利用したテクニカル売買戦略
小泉洋八; 鈴木智也, Corresponding
Technical Analysts Journal, Oct. 2015, [Reviewed] - 突発的な裁定機会を利用した共和分ペアトレーディング
鈴木智也; 成松優, Lead
Technical Analysts Journal, Oct. 2015, [Reviewed] - Nonlinear AR-DCC Portfolio Model Considering Liquidity of Imperfect Markets
Inose Satoshi; Tomoya Suzuki, Corresponding
Proceedings of International Symposium on Nonlinear Circuits and Signal Processing, 01 Mar. 2015, [Reviewed] - Nonlinear Time-varying AR-ARCH Model Based on Chaos Prediction Model
Hajime Onuma; Tomoya Suzuki, Corresponding
Proceedings of International Symposium on Nonlinear Circuits and Signal Processing, 01 Mar. 2015, [Reviewed] - Adaptive Optimization of Embedding Parameters by Minimizing Prediction Risk
Megumi Yokouchi; Tomoya Suzuki, Corresponding
Proceedings of International Symposium on Nonlinear Circuits and Signal Processing, 01 Mar. 2015, [Reviewed] - Technical Trading Strategy Using the Reaction to Financial Market Jumps
Hiroya Koizumi; Tomoya Suzuki, Corresponding
Proceedings of International Symposium on Nonlinear Circuits and Signal Processing, 01 Mar. 2015, [Reviewed] - Improving Predictive Power and Risk Reduction of the Portfolio Models Based on Principal Component Analysis
Kazuki Yanagisawa; Tomoya Suzuki, Corresponding
Proceedings of International Symposium on Nonlinear Circuits and Signal Processing, 01 Mar. 2015, [Reviewed] - 決定論的非線形予測に基づいた時空間テクニカル分析
鈴木智也; 林大賀, Lead
電子情報通信学会論文誌A, 01 Feb. 2015, [Reviewed] - Small-world Properties Evaluated by Exchanging Network topology
Tomoya Suzuki; Kuniaki Ohkura; Masayuki Okazawa, Lead, The present study quantified the degree of the small-world (SW) property defined by Watts, and evaluated its achievement level to characterize complex networks. However, because this process has a combinatorial optimization problem, we applied the chaos neural network (CNN) and the simulated annealing (SA), and confirmed their performance in terms of optimized values and numerical costs. Next, we visualized the original network and its optimized networks whose SW property was maximized or minimized by exchanging the original network topology. As a result, although CNN and SA require huge computational time, we confirmed that they can evaluate the SW property and even real SW networks still have plenty of room to enlarge their own SW property., WORLD SCIENTIFIC PUBL CO PTE LTD
International Journal of Modern Physics C,International Journal of Modern Physics C, 2015, [Reviewed] - Nonlinear Time-varying AR-ARCH Model Based on Chaos Prediction Model and its Statistical Significance Tests
Tomoya Suzuki; Hajime Onuma, Lead
Journal of Communication and Computer, 2015, [Reviewed] - Enhancing Predictive Power and Risk-reduction Efficiency of the Portfolio Models Based on Principal Component Analysis
Kazuki Yanagisawa; Tomoya Suzuki, Corresponding
Journal of Signal Processing, 2015, [Reviewed] - Minimizing Prediction Risk for Adaptive Optimization of Embedding Parameters for Noisy and Short Data
Megumi Yokouchi; Tomoya Suzuki, Corresponding
Journal of Signal Processing, 2015, [Reviewed] - Financial Technical Indicator Based on Chaotic Bagging Predictors for Adaptive Stock Selection in Japanese and American Markets
Tomoya Suzuki; Yushi Ohkura, Lead
Physica A, 2015, [Reviewed] - Mean-Variance Portfolio Model Modified by Nonlinear Bagging Predictors
Tomoya Suzuki; Kiyoharu Tanaka, In Markowitz s mean-variance portfolio model, the probability distribution of a future return is composed of recent historical prices, and the future return and future risk are estimated as the mean and standard deviation of the distribution, respectively. Namely, the future return is predicted by a simple moving average, and the risk is simply the historical fluctuation. In this study, to improve the prediction accuracy of the future return, we apply a nonlinear prediction method following local spatial dynamics, and to estimate the future risk, we produce a probability distribution by aggregating predicted values by the bagging algorithm. Then, each risk is reduced by making a portfolio, that is, we apply the portfolio effect. Namely, our method attempts to simultaneously improve the prediction accuracy and reduce the risk of its prediction error. To confirm the validity of our method, we performed investment simulations. As a result, we could realize higher profit and lower risk in investment than by the conventional method., Research Institute of Signal Processing, Japan
Journal of Signal Processing, 01 Nov. 2014, [Reviewed] - Combinatorial Optimization of Financial Technical Indicators Based on Bayesian Network
Haruaki Sakaki; Tomoya Suzuki
Proceedings of 2014 International Symposium on Nonlinear Theory and its Applications, Sep. 2014, [Reviewed] - Stock Portfolio Optimization Based on Nonlinear Prediction and DCC-GARCH Model
Inose Satoshi; Tomoya Suzuki; Kazuo Yamanaka
Proceedings of 2014 International Symposium on Nonlinear Theory and its Applications, Sep. 2014, [Reviewed] - Application of the Principal Components Analysis to the Nonlinear Portfolio Model
Kai Morimoto; Masahiro Saito; Satoshi Inose; Atsushi Kannari; Tomoya Suzuki
Journal of Signal Processing, 01 Jul. 2014, [Reviewed] - Long and Short Strategy Based on the Nonlinear DCC Portfolio Model
Satoshi Inose; Tomoya Suzuki; Kazuo Yamanaka:
Proceedings of International Symposium on Nonlinear Circuits and Signal Processing, Mar. 2014, [Reviewed] - Application of the Nonlinear Portfolio Model to Foreign Exchange Trading
Hirotake Wachi; Satoshi Inose; Tomoya Suzuki
Proceedings of International Symposium on Nonlinear Circuits and Signal Processing, Mar. 2014, [Reviewed] - Automated Trading System Using the Nonlinear Portfolio Model Implemented by Matlab and MetaTrader
Thanh Vu Tat; Satoshi Inose; Tomoya Suzuki
Proceedings of International Symposium on Nonlinear Circuits and Signal Processing, Mar. 2014 - Application of the Principal Components Analysis to the Nonlinear Portfolio Model
Kai Morimoto; Masahiro Saito; Satoshi Inose; Atsushi Kannari; Tomoya Suzuki
Proceedings of International Symposium on Nonlinear Circuits and Signal Processing, Mar. 2014, [Reviewed] - Machine learning of economic sensitive industries for domestic equity management
Proc. of RISP International Workshop on NCSP, Feb. 2014, [Reviewed] - Risk Reduction for Nonlinear Prediction and its Application to the Surrogate Data Test
Tomoya Suzuki; Kazuya Nakata, We propose a method for estimating nonlinear prediction risk using a bagging algorithm that involves ensemble learning. First we estimate the probability distribution of a future state as the ensemble set obtained using bagging predictors, and consider its standard deviation as the prediction risk. We can then improve the prediction reliability by avoiding dangerous predictions if the estimated prediction risk is high. As an application of this risk reduction method, we improve the power of surrogate data tests for system identification. Low prediction accuracy and poor system identification are caused by short and noisy data, so we perform simulations using short data derived from noisy chaotic models and real systems to confirm the validity of our method. (C) 2013 Elsevier B.V. All rights reserved., ELSEVIER SCIENCE BV
Physica D, 2014, [Reviewed] - Bipower Variation を用いた新しいテクニカル指標
山田雅章; 鈴木智也
テクニカルアナリストジャーナル, 2014, [Reviewed] - Tradeoff between Commission and Frequency of Rebalancing the Nonlinear Portfolio Model
Inose Satoshi; Tomoya Suzuki; Kazuo Yamanaka
Proceedings of International Symposium on Nonlinear Circuits and Signal Processing, Mar. 2013, [Reviewed] - Nonlinear Technical Analysis Using Spatial Historical Data
Yushi Ohkura; Tomoya Suzuki
Proceedings of International Symposium on Nonlinear Circuits and Signal Processing, Mar. 2013, [Reviewed] - Modified Bollinger Bands Based on Nonlinear Theory for Arbitrage Trading Strategies
Yusaku Hirano; Taiga Hayashi; Tomoya Suzuki
Proceedings of International Symposium on Nonlinear Circuits and Signal Processing, Mar. 2013, [Reviewed] - Portfolio Selection Based on Nonlinear Time Sereis Prediction
猪瀬悟史; 鈴木智也
電子情報通信学会論文誌A, 2013, [Reviewed] - Nonlinear Portfolio Model and its Rebalance Strategy
Inose Satoshi; Tomoya Suzuki; Kazuo Yamanaka, A nonlinear portfolio model was formulated by applying a nonlinear prediction method and its prediction error to the Markowitz mean-variance portfolio model. Also, the Sharpe ratio, which is a typical evaluation function of portfolio optimization, was modified to adopt stock-trading commissions and the trading-unit system, which are inevitable for portfolio rebalancing in real investment. Then, we discussed the best rebalancing frequency from the viewpoint of the trade-off between prediction accuracy and rebalancing costs. By investment simulations based on real stock data, we confirmed that shorter-term rebalancing is more effective even if we are required to pay higher commissions because short-term nonlinear prediction works better to estimate future return rates and to reduce investment risks., The Institute of Electronics, Information and Communication Engineers
Nonlinear Theory and Its Applications, IEICE, 2013, [Reviewed] - Dynamical Combinatorial Optimization for Predicting Multivariate Complex Systems
Tomoya Suzuki
Journal of Signal Processing, Dec. 2012, [Reviewed] - Stock Portfolio Management Based on Nonlinear Prediction Model
Inose Satoshi; Tomoya Suzuki; Kazuo Yamanaka
Proceedings of 2012 International Symposium on Nonlinear Theory and its Applications, Oct. 2012, [Reviewed] - New Bollinger Bands for Nonlinear Technical Analysis of Pairs Trading
Taiga Hayashi; Tomoya Suzuki
Proceedings of 2012 International Symposium on Nonlinear Theory and its Applications, Oct. 2012, [Reviewed] - Dynamical Portfolio Theory by Nonlinear Bagging Predictors
Kiyoharu Tanaka; Tomoya Suzuki
Proceedings of 2012 International Symposium on Nonlinear Theory and its Applications, Oct. 2012, [Reviewed] - Data Sampling Strategies for Nonlinear Analyses and Predictions of Deterministic Jump Systems
大塚 陽介; 鈴木 智也
情報処理学会論文誌 数理モデル化と応用, 01 Mar. 2012 - Appropriate Time Scales for Nonlinear Analyses of Deterministic Jump Systems
Tomoya Suzuki, In the real world, there are many phenomena that are derived from deterministic systems but which fluctuate with nonuniform time intervals. This paper discusses the appropriate time scales that can be applied to such systems to analyze their properties. The financial markets are an example of such systems wherein price movements fluctuate with nonuniform time intervals. However, it is common to apply uniform time scales such as 1-min data and 1-h data to study price movements. This paper examines the validity of such time scales by using surrogate data tests to ascertain whether the deterministic properties of the original system can be identified from uniform sampled data. The results show that uniform time samplings are often inappropriate for nonlinear analyses. However, for other systems such as neural spikes and Internet traffic packets, which produce similar outputs, uniform time samplings are quite effective in extracting the system properties. Nevertheless, uniform samplings often generate overlapping data, which can cause false rejections of surrogate data tests., AMER PHYSICAL SOC
Physical Review E, 09 Jun. 2011, [Reviewed] - Analysis on the Efficiency of Statistical Measures to Identify Network Structure of Chaos Coupled Systems
Yuta Ueoka; Tomoya Suzuki; Seiichi Yamamoto, Real systems often show complex behavior due to interaction among many elements composing a large-scale network. To model and predict these systems, it is desired to estimate network structures by using only time-series data observed as behavior of systems. Although several kinds of estimation techniques have been proposed, the optimum technique might be different according to properties of systems. To analyze the possibility, we estimate interactions of chaos coupled systems by four typical types of estimation techniques. For numerical simulations, we adopt the coupled map lattice, which is a model of large-scale complex systems, and we modify it so as to control the degree of synchronization and the instability of systems by changing the coupling strength and the topology of interaction among elements. As results, we can confirm that the optimum technique depends on properties of system, and then we clarify the reason from the viewpoint of synchronization and the Lyapunov exponents. Moreover, as an application, we predict future behavior of each element with new prediction model based on estimated interactions, and we demonstrate the efficiency of this prediction method., WORLD SCIENTIFIC PUBL CO PTE LTD
International Journal of Modern Physics C, 27 Aug. 2010, [Reviewed] - Estimating Structure of Multivariate Systems with Genetic Algorisms for Nonlinear Prediction
Tomoya Suzuki; Yuta Ueoka; Haruki Sato, Although we can often observe time-series data of many elements, these elements do not always interact with each other. This paper proposes a scheme to estimate the interdependency among observed elements only by time-series data, which is useful for selecting essential elements to optimize multivariate prediction model. Because this estimation is a sort of combinatorial optimization problems, we applied the genetic algorithm as a method to moderate this problem. Through some simulations, we confirmed performance of our method, which can identify interaction of multivariate system and can improve its prediction accuracy. Especially, our method can be applied to predict real foreign-exchange markets even if system has nonstational property and its structure changes dynamically., AMER PHYSICAL SOC
Physical Review E, 07 Dec. 2009, [Reviewed] - 情報伝達に基づいた有向重み付き複雑ネットワーク解析
鈴木智也
情報処理学会論文誌 数理モデル化と応用, 2009, [Reviewed] - 複雑システムにおけるネットワーク中心性が予測精度に与える影響
鈴木智也; 池田真一
情報処理学会論文誌 数理モデル化と応用, 2009, [Reviewed] - 時系列データの天底予測のための非線形予測法
鈴木智也; 太田真喜
情報処理学会論文誌 数理モデル化と応用, 2009, [Reviewed] - Application of Chaos Game Representation to Nonlinear Time Series Analysis
Tomoya Suzuki; Tohru Ikeguchi; Masuo Suzuki
Fractals, 2007, [Reviewed] - Bootstrap Nonlinear Prediction
Daisuke Haraki; Tomoya Suzuki; Hiroki Hashiguchi; Tohru Ikeguchi, Estimating the Jacobian matrix of a nonlinear dynamical system through observed time-series data is one of the important steps in predicting future states of the time series. The Jacobian matrix is estimated using local information about divergences of nearby trajectories. Although the basic algorithm for estimating the Jacobian matrix generally works well, it often fails for short or noisy data series. In this paper, we proposed a scheme to effectively use near-neighbor information for more accurate estimation of the Jacobian matrix using the bootstrap resampling method. Then, to confirm the validity of the proposed method, we applied it to a mathematical model and several real time series. As a result, we confirmed that the proposed method greatly improves nonlinear predictability, not only for noise-corrupted mathematical models but also for real time series., AMERICAN PHYSICAL SOC
Physical Review E, 2007 - Algorithms for Generating Surrogate Data for Sparsely Quantized Time Series
Tomoya Suzuki; Tohru Ikeguchi; Masuo Suzuki, The method of surrogate data is frequently used for a statistical examination of nonlinear properties underlying original data. If surrogate data sets are generated by a null hypothesis that the data are derived by a linear process, a rejection of the hypothesis means that the original data have more complex properties. However, we found that if an algorithm for generating surrogate data, for example, amplitude adjusted Fourier transformed, is applied to sparsely quantized data, there are large discrepancies between their power spectrum and that of the original data in lower frequency regions. We performed some simulations to confirm that these errors often lead to false rejections.
In this paper, in order to prevent such drawbacks, we advance an extended hypothesis, and propose two improved algorithms for generating surrogate data that reduce the discrepancies of the power spectra. We also confirm the validity of the two improved algorithms with numerical simulations by showing that the extended null hypothesis can be rejected if the time series is produced from chaotic dynamical systems. Finally, we applied these algorithms for analyzing financial tick data as a real example; then we showed that the extended null hypothesis cannot be rejected because the nonlinear statistics or nonlinear prediction errors exhibited are the same as those of the original financial tick time series. (C) 2007 Elsevier B.V. All rights reserved., ELSEVIER SCIENCE BV
Physica D, 2007, [Reviewed] - Evaluating nonlinearity and validity of nonlinear modeling for complex time series
Tomoya Suzuki; Tohru Ikeguchi; Masuo Suzuki, Even if an original time series exhibits nonlinearity, it is not always effective to approximate the time series by a nonlinear model because such nonlinear models have high complexity from the viewpoint of information criteria. Therefore, we propose two measures to evaluate both the nonlinearity of a time series and validity of nonlinear modeling applied to it by nonlinear predictability and information criteria. Through numerical simulations, we confirm that the proposed measures effectively detect the nonlinearity of an observed time series and evaluate the validity of the nonlinear model. The measures are also robust against observational noises. We also analyze some real time series: the difference of the number of chickenpox and measles patients, the number of sunspots, five Japanese vowels, and the chaotic laser. We can confirm that the nonlinear model is effective for the Japanese vowel /a/, the difference of the number of measles patients, and the chaotic laser., AMER PHYSICAL SOC
Physical Review E, 2007, [Reviewed] - Bootstrap prediction intervals for nonlinear time-series
Daisuke Haraki; Tomoya Suzuki; Tohru Ikeguchi, To evaluate predictability of complex behavior produced from nonlinear dynamical systems, we often use normalized root mean square error, which is suitable to evaluate errors between true points and predicted points. However, it is also important to estimate prediction intervals, where the future point will be included. Although estimation of prediction intervals is conventionally realized by an ensemble prediction, we applied the bootstrap resampling scheme to evaluate prediction intervals of nonlinear time-series. By several numerical simulations, we show that the bootstrap method is effective to estimate prediction intervals for nonlinear time-series., SPRINGER-VERLAG BERLIN
INTELLIGENT DATA ENGINEERING AND AUTOMATED LEARNING - IDEAL 2006, PROCEEDINGS, 2006, [Reviewed] - Bootstrap Prediction Intervals for Nonlinear Time-Series
Daisuke Haraki; Tomoya Suzuki; Tohru Ikeguchi, To evaluate predictability of complex behavior produced from nonlinear dynamical systems, we often use normalized root mean square error, which is suitable to evaluate errors between true points and predicted points. However, it is also important to estimate prediction intervals, where the future point will be included. Although estimation of prediction intervals is conventionally realized by an ensemble prediction, we applied the bootstrap resampling scheme to evaluate prediction intervals of nonlinear time-series. By several numerical simulations, we show that the bootstrap method is effective to estimate prediction intervals for nonlinear time-series., SPRINGER-VERLAG BERLIN
Proceedings of 7th International Conference on Intelligent Data Engineering and Automated Learning, 2006, [Reviewed] - Effects of data windows on the methods of surrogate data
Tomoya Suzuki; Tohru Ikeguchi; Masuo Suzuki, To generate surrogate data in nonlinear time series analysis, the Fourier transform is generally used. In the calculation of the Fourier transform, the time series is assumed to be periodic. Because such an assumption does not always hold true, the estimation accuracy of the Fourier transformed data and thus the power spectra is reduced. Due to such an estimation error, it is also possible that the surrogate test will lead to a false conclusion
for example, that a linear time series is nonlinear. In this paper, we experimentally evaluated the effects of data windows from the viewpoint of false rejections with several types of surrogate data. Our results indicate that if the data length becomes shorter, the false rejections by the data windows are reduced to a greater extent. However, if the data length is sufficient, the use of data windows is not a viable option. In the worst possible case wherein the linear memory of the original data is very long as in the nonstationary case, the critical length of the data for which the data windows were effective was approximately 1000. © 2005 The American Physical Society.
Physical Review E - Statistical, Nonlinear, and Soft Matter Physics, May 2005, [Reviewed] - Effect of Data Windows on the Models of Surrogate Data
Tomoya Suzuki; Tohru Ikeguchi; Masuo Suzuki, To generate surrogate data in nonlinear time series analysis, the Fourier transform is generally used. In the calculation of the Fourier transform, the time series is assumed to be periodic. Because such an assumption does not always hold true, the estimation accuracy of the Fourier transformed data and thus the power spectra is reduced. Due to such an estimation error, it is also possible that the surrogate test will lead to a false conclusion; for example, that a linear time series is nonlinear. In this paper, we experimentally evaluated the effects of data windows from the viewpoint of false rejections with several types of surrogate data. Our results indicate that if the data length becomes shorter, the false rejections by the data windows are reduced to a greater extent. However, if the data length is sufficient, the use of data windows is not a viable option. In the worst possible case wherein the linear memory of the original data is very long as in the nonstationary case, the critical length of the data for which the data windows were effective was approximately 1000., AMER PHYSICAL SOC
Physical Review E, 2005, [Reviewed] - A Model of Complex Behavior of Interbank Exchange Markets
Tomoya Suzuki; Tohru Ikeguchi; Masuo Suzuki, In the present paper, we analyze the complex interaction among three macroscopic variables, dealing time intervals, spreads between ask and bid prices and price movements, observed in actual interbank exchange markets. For this analysis, we propose a new model of interbank exchange dealings as a statistical system integrated by many dealers' actions with the methods of statistical physics. For evaluating the plausibility of our model, we compare outputs from the proposed model with the real data by reconstructing a state space with the above three variables, observing ensemble behavior in each day and estimating statistical properties. As a result, we can confirm that our model is plausible, and we perform the above analysis with our model from the viewpoint of statistical physics. (C) 2004 Elsevier B.V. All rights reserved., ELSEVIER SCIENCE BV
Physica A, 2004, [Reviewed] - Multivariable Nonlinear Analysis of Foreign Exchange Rates
Tomoya Suzuki; Tohru Ikeguchi; Masuo Suzuki, We analyze the multivariable time series of foreign exchange rates. These are price movements that have often been analyzed, and dealing time intervals and spreads between bid and ask prices. Considering dealing time intervals as event timing such as neurons' firings, we use raster plots (RPs) and peri-stimulus time histograms (PSTHs) which are popular methods in the field of neurophysiology. Introducing special processings to obtaining RPs and PSTHs time histograms for analyzing exchange rates time series, we discover that there exists dynamical interaction among three variables. We also find that adopting multivariables leads to improvements of prediction accuracy. (C) 2003 Elsevier Science B.V. All rights reserved., ELSEVIER SCIENCE BV
Physica A, 2003, [Reviewed] - On Evaluation Noise Levels for Quantized Observed Data
Tomoya Suzuki; Tohru Ikeguchi; Masuo Suzuki, In recent years, quantitative methods for evaluating chaotic properties have been developed in the field of nonlinear time-series analysis. The embedding theorem, which is a mathematical background for the methods, assumes an ideal situation in which noiseless time series are observed with infinite resolution and an infinite amount of data points. However, under real situations we cannot ignore two classes of noises which are included in really observed data. The first one is observational and dynamical noises which depend on internal nonlinear systems and performance of observational instruments. The second one is quantization error included in the time series since we usually use digital computers for applying the methods. In the present paper, we derive formulae for evaluating the levels of observational and quantization noises in the case of embedding observed time series in reconstructed state spaces. By measuring a distance between noiseless and noisy attractors, we also confirm that the derived formulae are appropriate for quantifying the noise included in the reconstructed attractor., WORLD SCIENTIFIC PUBL CO PTE LTD
International Journal of Modern Physics C, 2003, [Reviewed]
MISC
- 金融におけるテキストマイニングと機械学習応用
鈴木智也; 中川慧; 伊藤友貴; 坂地泰紀
人工知能学会誌, 01 Jun. 2021
Lead - 人工知能の集合知によるアルゴリズム運用
鈴木智也
テクニカルアナリストジャーナル, 2018
Lead - ニューラルネットワークの集団学習による価格変動パターンの自動検出および自信度の評価
鈴木智也
テクニカルアナリストジャーナル, 2016
Lead - 時空間決定論的テクニカル分析
鈴木智也
テクニカルアナリストジャーナル, 2014 - Stock Portfolio Management with Nonlinear Time Series Prediction
Inose Satoshi; Tomoya Suzuki
Proceedings of International Symposium on Nonlinear Circuits and Signal Processing, Mar. 2012 - Data Sampling Strategies for Long-Term Predictions of Deterministic Jump Systems
Yousuke Otsuka; Tomoya Suzuki
Proceedings of International Symposium on Nonlinear Circuits and Signal Processing, Mar. 2012 - Evaluating the Risk of Nonlinear Prediction with the Bagging Algorithm
Kazuya Nakata; Tomoya Suzuki
Proceedings of International Symposium on Nonlinear Circuits and Signal Processing, Mar. 2012 - Self-organizing small-world structure of neural networks by STDP learing rule
Tomoya Suzuki; Tohru Ikeguchi
同志社大学理工学研究所 研究報告, 2009, [Reviewed] - Combinatorial Optimization for Multivariate Nonlinear Prediction
Tomoya Suzuki; Yuta Ueoka; Haruki Sato
Proceedings of International Symposium on Nonlinear Circuits and Signal Processing, 2009, [Reviewed] - Dynamical Optimization for Nonlinear Prediction
Shougo Kaneko; Tomoya Suzuki
Proceedings of International Symposium on Nonlinear Circuits and Signal Processing, 2009 - Available Partial Information to Estimate the Whole Structure of Complex Systems
Yuta Ueoka; Tomoya Suzuki; Seiichi Yamamoto
Proceedings of International Symposium on Nonlinear Circuits and Signal Processing, 2009, [Reviewed] - Efficiency of Statistical Measures to Estimate Network Structure of Chaos Coupled Systems
Yuta Ueoka; Tomoya Suzuki; Tohru Ikeguchi; Yoshihiko Horio
Proceedings of 2008 International Symposium on Nonlinear Theory and its Applications, 2008, [Reviewed] - 多変量データを用いた複雑ネットワーク構造の推定と時系列予測への応用
鈴木智也
同志社大学理工学研究所 研究所報, 2008 - 多変量観測時系列データからの複雑ネットワーク構造の推定:経済市場や神経ネットワークを例に
同志社大学理工学研究所 研究所報, 2007 - Bootstrap Prediction Intervals for Nonlinear Time Series
Daisuke Haraki; Tomoya Suzuki; Tohru Ikeguchi
Lecture Notes in Computer Science, 2006, [Reviewed] - Nonlinear Analysis of the Pollen Scattering Data
Kenichi Aikawa; Tomoya Suzuki; Tohru Ikeguchi
Proceedings of International Symposium on Nonlinear Circuits and Signal Processing, 2005 - Transition from Random to Small-World Neural Networks by STDP Learning Rule
Tomoya Suzuki; Tohru Ikeguchi
Proceedings of International Symposium on Nonlinear Circuits and Signal Processing, 2005, [Reviewed] - Bootstrap Estimates for Nonlinear Predictors
Daisuke Haraki; Tomoya Suzuki; Tohru Ikeguchi
Proceedings of International Symposium on Nonlinear Theory and its Applications, 2005, [Reviewed] - A Measure for Nonlinear Predictability and Information Criteria
Tomoya Suzuki; Tohru Ikeguchi; Masuo Suzuki
Proceedings of International Symposium on Nonlinear Circuits and Signal Processing, 2004, [Reviewed] - Multivariable Modeling of Complex Behavior of Foreign Exchange Market
Tomoya Suzuki; Tohru Ikeguchi; Masuo Suzuki
Toward Control of Economic Change: Application of Econophysics, 2003, [Reviewed] - A Novel Model for Foreign Exchange Markets
Tomoya Suzuki; Tohru Ikeguchi; Masuo Suzuki
Proceedings of the Second Nikkei Econophysics Research Workshop and Symposium, 2002, [Reviewed] - Nonlinear Analysis on Interspike Interval Time Series from Foreign Exchange Rates
SUZUKI T.
Proceedings of 2001 International Symposium on Nonlinear Theory and its Applications, 2001, [Reviewed]
Lead
Books and other publications
Lectures, oral presentations, etc.
- 機械学習による企業の自社株買い行動の予測可能性
鈴木智也; 望月孝太郎; 田村空生; 加唐丈裕
情報処理学会知能システム研究会, 26 Mar. 2024 - パネルディスカッション・理工系って何するの?
理工系進路選択応援シンポジウム, 18 Feb. 2024, [Invited] - 機械学習を用いた車両部品の故障判断
佐野龍太郎; 工藤大輝; 鈴木智也
電子情報通信学会総合大会, Jan. 2024 - 銘柄固有リターンによるクロスセクション型テクニカル分析
圷智大; 鈴木隆司; 鈴木智也
電子情報通信学会総合大会, Jan. 2024 - ゴトウビアノマリーの周知によって発生し得るFX売買戦略
別所宏紀; 杉本誠忠; 鈴木智也
複雑コミュニケーションサイエンス研究会, 17 Nov. 2023 - 複数時間スケールのファインチューニングによる金融機械学習の精度向上
複雑コミュニケーションサイエンス研究会, 17 Nov. 2023 - 景気敏感業種を考慮した国内株式運用の機械学習
織田望夢; 鈴木隆司; 鈴木智也
日本機械学会茨城講演会, Aug. 2023 - 企業財務業績に基づくカスタマーモメンタムの有効性検証
新澤和弥; 鈴木智也
日本機械学会茨城講演会, Aug. 2023 - 銘柄間の群集心理を可視化するクロスセクション型テクニカル分析
圷智大; 鈴木隆司; 鈴木智也
日本機械学会茨城講演会, Aug. 2023 - 国際株式運用における多目的最適化支援ツール
澤畑英介; 塙祥傳; 川又仁通; 鈴木智也
日本機械学会茨城講演会, Aug. 2023 - ゴトウビアノマリーにおけるEBS注文板情報を用いた売買行動分析
別所宏紀; 杉本誠忠; 鈴木智也
日本機械学会茨城講演会, 01 Aug. 2023 - テキストマイニングによるテーマ型ファンドの組成支援
駱子傑; 戴子儀; 倉本渉; 鈴木智也
日本機械学会茨城講演会, 01 Aug. 2023 - 日米サプライチェーンにおけるカスタマーモメンタムとサプライヤー特性の交互作用
酒井優樹; 関谷健; 谷塚智成; 鈴木智也
日本機械学会茨城講演会, 01 Aug. 2023 - 常陽銀行×茨城大 共同研究の中間報告会
常陽銀行×茨城大 共同研究の中間報告会, 17 May 2023, [Invited]
20230517 - テーマ型ファンド組入銘柄の選定理由の可視化
戴子儀; 駱子傑; 倉本渉; 鈴木智也
電子情報通信学会総合大会, 09 Mar. 2023 - 英字テキスト解析によるテーマ型ファンドの自動銘柄選定
佐藤瑠星; 佐野龍太郎; 倉本渉; 鈴木智也
電子情報通信学会総合大会, 09 Mar. 2023 - 外国為替レートの予測可能性に関する要因分析
電子情報通信学会総合大会, 09 Mar. 2023 - サプライチェーン上の位置がカスタマーモメンタムに与える影響
関谷健; 酒井優樹; 谷塚智成; 鈴木智也
電子情報通信学会総合大会, 09 Mar. 2023 - Forex Trading Strategy That Might Be Executed Due to the Popularity of Gotobi Anomaly
Hiroki Bessho; Takanari Sugimoto; Tomoya Suzuki
International Workshop on Nonlinear Circuits, Communications and Signal Processing, 02 Mar. 2023 - Long-Term Modeling of Financial Machine Learning with Multiple Time Scales
Kazuki Amagai; Riku Tanaka; Tomoya Suzuki
International Workshop on Nonlinear Circuits, Communications and Signal Processing, 02 Mar. 2023 - Composition of Thematic Equity Funds by Searching Multi and Unknown Words
International Workshop on Nonlinear Circuits, Communications and Signal Processing, 02 Mar. 2023 - 「今こそ仲値を科学する」鈴木智也 × Trader Kaibe
雑誌「外国為替」, 22 Feb. 2023, [Invited] - 人間とAI、為替取引においてはどちらが優位になるのか?
有限会社グリーン・アース, 05 Dec. 2022, [Invited] - 金融業務におけるAI・データサイエンスの活用事例
日本塑性加工学会 塑性加工技術フォーラム, 02 Dec. 2022, [Invited] - 複数時間軸情報を用いたオートエンコーダーによる行動経済学的特性の抽出
川田瑛貴; 雨谷暦樹; 田中 陸; 鈴木智也
人工知能学会金融情報学研究会, 12 Mar. 2022 - FX市場におけるオートエンコーダの異常検知を活用したポートフォリオ運用
雨谷暦樹; 川田瑛貴; 田中 陸; 鈴木智也
人工知能学会金融情報学研究会, 12 Mar. 2022 - 機械学習による理論株価の評価および投資家心理の抽出
塚原悠輝; 田中 陸; 鈴木智也
電子情報通信学会CCS研究会, 18 Nov. 2021 - 機械学習による為替フォワード取引期間の判別モデル
雉子波晶; 杉本誠忠; 酒本隆太; 鈴木智也
人工知能学会金融情報学研究会, 09 Oct. 2021 - 国内株式投資信託における資金フローの非線形モデリング
吉田遼平; 中道拓馬; 田中 陸; 鈴木智也
人工知能学会金融情報学研究会, 09 Oct. 2021 - 投資信託の運用パフォーマンスと資金フローの非線形関係
中道拓馬; 吉田遼平; 田中 陸; 鈴木智也
電子情報通信学会ソサイエティ大会, 16 Sep. 2021 - EBS板情報を用いた外国為替レートの短期予測
山口風樹; 杉本誠忠; 酒本隆太; 鈴木智也
電子情報通信学会ソサイエティ大会, 16 Sep. 2021 - オートエンコーダの異常検知による行動経済学的特性の抽出
電子情報通信学会総合大会, 10 Mar. 2021 - 国内投資信託における資金流出入要因の極性分析
電子情報通信学会総合大会, 10 Mar. 2021 - 機械学習による全国中古車オークション会場の割安特性
電子情報通信学会総合大会, 10 Mar. 2021 - 機械学習による中古車落札価格の要因分析及び異常検知
情報処理学会MPS研究会, 20 Dec. 2020 - ニュースヘッドラインの機械学習によるアクティブ運用
JAFEE大会, 20 Aug. 2020 - カバー付き金利平価からの乖離を利用した為替フォワード取引
JAFEE大会, 20 Aug. 2020 - 円相場における日本特有のゴトウビアノマリー
JAFEE大会, 20 Aug. 2020 - 金融ニュースを用いた個別銘柄の状況の類似度による可視化
日本機械学会茨城講演会, 10 Aug. 2020 - Word2Vec を用いたニューステキストの ESG ファクター運用
人工知能学会全国大会, 10 Jun. 2020 - 為替フォワード取引における最適タイミングの機械学習
人工知能学会全国大会, 10 Jun. 2020 - 暗号資産における共和分ペアトレード
人工知能学会全国大会, 10 Jun. 2020 - 中古車状態の機械学習による落札価格の推定
電子情報通信学会総合大会, 18 Mar. 2020 - 異常検知による中古車落札価格の割安・割高判断
電子情報通信学会総合大会, 17 Mar. 2020 - 中古車の特徴量が落札価格へ及ぼす影響分析
電子情報通信学会総合大会, 17 Mar. 2020 - ニュース重要単語の機械学習によるアクティブ運用
人工知能学会金融情報学研究会, 14 Mar. 2020 - ニューステキストを用いたESGファクター運用
人工知能学会金融情報学研究会, 14 Mar. 2020 - AIってなに? 誰にもわかる! 人工知能の話
熟年ネット・ひたちセミナー, 13 Feb. 2020, J-net, [Invited] - AIにより機械化が進む資産運用ビジネス
ニッキン投信情報, 14 Jan. 2020, [Invited] - 実務における人工知能AIの可能性と限界
型技術ワークショップ, 28 Nov. 2019, [Invited] - 中古車の落札価格に伴う特徴量の影響度分析
工藤大輝; 福西亮介; 黛広樹; 鈴木智也
日本機械学会茨城講演会, Aug. 2019 - 金融ニューステキストを用いた気配情報の可視化
梅津信幸; 廣川優樹; 鈴木智也; 江口潤一
日本機械学会茨城講演会, Aug. 2019 - BERTモデルとニュースヘッドラインによる AI 運用システムの試作
史文ガイ; 細木唯以; 三好勝博; 江口潤一; 佐々木稔; 鈴木智也
日本機械学会茨城講演会, Aug. 2019 - AIと資産運用 〜 人工知能AIはどこまで資産運用に役立つか
Ai・ICT次世代広域応用教育研究センターセミナー, 17 Jul. 2019, [Invited] - パネルディスカッション,AI・フィンテックによる運用の未来
日本テクニカルアナリスト協会, 12 Jul. 2019, 日本テクニカルアナリスト協会, [Invited] - パネルディスカッション AI・フィンテックによる運用の未来
日本テクニカルアナリスト協会, 12 Jul. 2019, 日本テクニカルアナリスト協会, [Invited] - 人工知能の夢と現実 〜 AlphaGo, AI運用, 無くなる仕事, 人間は支配されるか?
GIS総合研究所, 19 Jun. 2019, GIS総合研究所, [Invited] - パネルディスカッション,AIと資産運用 〜 人工知能AIはどこまで資産運用に役立つか
日本金融学会, 26 May 2019, [Invited] - パネルディスカッション AIと資産運用 〜 人工知能AIはどこまで資産運用に役立つか
日本金融学会, 26 May 2019, [Invited] - 人工知能AIによる地域創生, 期待されるイノベーション
産学連携講演会, 18 Mar. 2019, 福島県白河市, [Invited] - 人工知能AIはどこまで資産運用に役立つか
日本CFA協会セミナー, 13 Mar. 2019, 日本CFA協会, [Invited] - 外国為替市場におけるゴトウビアノマリーの有用性検証
秋山朋也; 塚瀬正人; 鈴木恒平; 鈴木智也
電子情報通信学会総合大会, Mar. 2019 - ニュースヘッドラインの機械学習による投資判断
電子情報通信学会総合大会, Mar. 2019 - 人工知能AIの正体を知り, ビジネスに活かすヒントを探る
地方創生セミナー, 05 Feb. 2019, NTTドコモ/茨城新聞社/茨城県, [Invited] - AIの集合知によるX-Techと産学連携機能強化
学長学術表彰記念講演会, 10 Dec. 2018, 茨城大学, [Invited] - 人工知能の夢と現実 〜 Alは投資に役立つのか?
投資日報社セミナー, 27 Oct. 2018, 投資日報社, [Invited] - 深層学習によるオートオークション落札価格予測
櫻井大宙; 工藤大輝; 長谷川恵理子; 下山力三; 福西亮介; 黛広樹; 鈴木智也
電子情報通信学会 非線形問題研究会, Oct. 2018 - カバー先銀行の集合知による外国為替レート予測
鈴木丈裕; 鈴木智也
人工知能学会 ビジネスインフォマティクス研究会, Sep. 2018 - 機械学習によるオートオークション落札価格の予測
人工知能学会 ビジネスインフォマティクス研究会, Sep. 2018 - AI・機械学習の現実的な利活用を冷静に考える
茨城県情報通信産業支援協議会, 25 Jun. 2018, 茨城県, [Invited] - カバー先銀行の建値情報を用いた外国為替市場の価格予測: プロ集団による集合知の活用
鈴木丈裕; 鈴木智也
電子情報通信学会 複雑コミュニケーションサイエンス研究会, Jun. 2018 - 顧客の取引履歴情報を用いた外国為替市場の価格予測: アマチュア集団による集合知の活用
矢野和洞; 鈴木智也
電子情報通信学会 複雑コミュニケーションサイエンス研究会, Jun. 2018 - 人工知能の夢と現実 〜 AlphaGo, AI運用, 無くなる仕事, 人間は支配されるか?
茨城県庁ITセミナー, 01 Feb. 2018, 茨城県, [Invited] - 人工知能AIに関する情報整理と今後の可能性
埼玉県環境計量協議会, 26 Jan. 2018, 埼玉県環境計量協議会, [Invited] - AI運用に関するパネルディスカッション
大和証券セミナー, 12 Jan. 2018, 大和証券, [Invited] - AIアルゴリズム運用の可能性
茨城県学生ビジネスコンテスト, 23 Nov. 2017, 茨城県, [Invited] - Collective Artificial Intelligence for Mechanical Technical Analysis
Tomoya Suzuki
The IFTA 2017 Annual Conference, Oct. 2017, [Invited] - AIによる客観的テクニカル分析と問題点
大和証券セミナー, 31 Jul. 2017, 大和証券, [Invited] - AlphaGo(囲碁プログラム)から人工知能と経営戦略の接点を探る
パートナー企業交流会, 28 Jun. 2017, 茨城大学, [Invited] - J-REIT市場における季節性分析
張明新; 中村貴司; 鈴木智也
電子情報通信学会複雑コミュニケーションサイエンス研究会, Jun. 2017 - ボラティリティ指標による金融市場のジャンプ検出および直後の反応
鶴田季丸; 鈴木智也
電子情報通信学会 複雑コミュニケーションサイエンス研究会, Jun. 2017 - 人工知能の集合知による機械的テクニカル戦略 〜コンセンサスレシオによる動的銘柄選択〜
IFTAジョン・ブルークス賞受賞記念講演会, 31 May 2017, 日本テクニカルアナリスト協会, [Invited] - 人工知能やFinTechに関する歴史と最新動向
年金資産運用研究会, 29 May 2017, 年金資産運用研究会, [Invited] - オートエンコーダによる金融市場のジャンプ検出および直後の反動
鈴木智也; 後藤弘行; 鶴田季丸; 小泉洋八; 神成敦
2016年度人工知能学会全国大会, Jun. 2016 - Ensemble Neural Networks for Identifying Market Patterns and their Confidence
Tomoya Suzuki
The IFTA 2015 Annual Conference, 30 Oct. 2015, [Invited] - 非線形ポートフォリオモデルを用いた外国為替自動取引システムの構築
和知宏武; Vu Tat Thanh; 猪瀬悟史; 神成敦; 鈴木智也
電子情報通信学会 非線形問題研究会, Jan. 2014 - DCCモデルを適用した非線形ポートフォリオモデルによるロングショート戦略
猪瀬悟史; 鈴木智也; 山中一雄
電子情報通信学会 非線形問題研究会, Jan. 2014 - Spatiotemporal Technical Analyses based on Deterministic Prediction Theory
Tomoya Suzuki
The IFTA 2013 Annual Conference, Oct. 2013, The International Federation of Technical Analysts (IFTA) - バギングによる平均分散ポートフォリオモデル
田中清春; 鈴木智也
電子情報通信学会2011年総合大会, Mar. 2012 - ペアトレーディングにおける非線形テクニカル分析
林大賀; 鈴木智也
電子情報通信学会2011年総合大会, Mar. 2012 - 金融工学における非線形時系列モデリング
鈴木智也; 猪瀬悟史; 田中清春; 林大賀; 大倉佑嗣
電子情報通信学会2011年総合大会, Mar. 2012 - 投資持続時間とリスクを考慮した合理的な手仕舞い戦略 ~ 損小利大戦略は最適か?
水野翔太; 鈴木智也
日本物理学会2011年次大会, Mar. 2012 - バギング型非線形予測による平均分散ポートフォリオモデル
鈴木智也; 猪瀬悟史; 田中清春
日本物理学会2011年次大会, Mar. 2012 - バギングによる非線形予測のリスク評価
仲田和也; 鈴木智也
電子情報通信学会 非線形問題研究会, 20 Oct. 2011 - カオスニューラルネットワークによる多目的最適化
岡澤政幸; 鈴木智也
電子情報通信学会 非線形問題研究会, 20 Oct. 2011 - 非線形時系列予測による株式ポートフォリオの運用
猪瀬悟史; 鈴木智也
電子情報通信学会 非線形問題研究会, 20 Oct. 2011 - ポートフォリオ構築問題における時系列予測モデルの活用
猪瀬悟史; 鈴木智也
情報処理学会 数理モデル化と問題解決研究会, 16 Sep. 2011 - 決定論的ジャンプ過程のシステム同定と長期予測に適したサンプリング手法の検討
大塚陽介; 鈴木智也
情報処理学会 数理モデル化と問題解決研究会, 16 Sep. 2011 - 決定論的ジャンプ過程のシステム同定と長期予測
鈴木智也; 大塚陽介
日本物理学会2010年次大会, 27 Mar. 2011 - カオス結合系の挙動と構造同定手法の関係
上岡祐太; 鈴木智也; 山本誠一
日本物理学会2010年次大会, 26 Mar. 2011 - 複雑システムの構造推定のためのグレンジャー因果性に基づく閾値決定法
上岡祐太; 鈴木智也; 山本誠一
情報処理学会2010年全国大会, 04 Mar. 2011 - 非線形予測誤差に基づいた株式ポートフォリオの構築
猪瀬悟史; 鈴木智也
情報処理学会2010年全国大会, 04 Mar. 2011 - 多目的組合せ最適化問題におけるカオスニューラルネットワークの性能評価
岡澤政幸; 鈴木智也
情報処理学会2010年全国大会, 04 Mar. 2011 - 決定論的ジャンプ過程の長期予測に適したデータサンプリング手法の検討
大塚陽介; 鈴木智也
情報処理学会2010年全国大会, 04 Mar. 2011 - バギングを用いた非線形時系列予測のリスク評価
仲田和也; 鈴木智也
情報処理学会2010年全国大会, 04 Mar. 2011 - 等時間間隔サンプリングによって見失う非線形システムの特徴
鈴木智也
情報処理学会2010年全国大会, Mar. 2010 - 等時間間隔サンプリングによって欠落するシステムの非線形性
鈴木智也
電子情報通信学会2009年総合大会, Mar. 2010 - 複雑システムの理解と予測のための観測時系列データの最適利用
鈴木智也
第52回自動制御連合講演会, 2009 - 非線形経済予測モデルの動的最適化
鈴木智也; 佐藤春樹; 金子彰吾
日本物理学会2009年次大会, 2009 - 非線形時系列解析における欠損データが及ぼす影響
三井貴視; 瀬木宏; 鈴木智也
電子情報通信学会2009年総合大会, 2009 - 多変量予測モデル構築における組合せ最適化問題
鈴木智也; 上岡祐太; 佐藤春樹; 金子彰吾
電子情報通信学会2009年総合大会, 2009 - 欠損を含む時系列データの非線形予測
瀬木宏; 三井貴視; 鈴木智也
情報処理学会2009年全国大会, 2009 - ブートストラップ法を用いた少数データに対する局所線形近似法
上野佑輔; 鈴木智也
情報処理学会2009年全国大会, 2009 - 部分的ネットワーク情報を利用した大域的ネットワーク構造の推定
上岡祐太; 鈴木智也; 山本誠一
情報処理学会2009年全国大会, 2009 - 進化的計算手法を用いた多変量予測モデルの動的最適化
佐藤春樹; 鈴木智也
情報処理学会2009年全国大会, 2009 - 進化的計算手法を用いた多変量システムの因果推定と予測問題への応用
鈴木智也; 上岡祐太; 佐藤春樹
情報処理学会2009年全国大会, 2009 - Characterizing Cluster Coefficient in Directed and Weighted Complex Networks on the Baisis of Information Flow
Tomoya Suzuki
Proceedings of 2008 International IEEE Workshop on Nonlinear Dynamics of Electronic Systems, 2008 - Estimating Network Structure of Chaos Coupled Systems
Yuta Ueoka; Tomoya Suzuki; Tohru Ikeguchi; Yoshihiko Horio
International IEEE Workshop on Nonlinear Dynamics of Electronic Systems, 2008 - ささやき声の非線形解析
鈴木智也; 池上亜由子
電子情報通信学会2008年総合大会, 2008 - カオス結合系におけるネットワークの推定
上岡裕太; 鈴木智也; 池口徹; 堀尾喜彦
電子情報通信学会2008年総合大会, 2008 - 局所線形近似法における予測アルゴリズムの改良
寺西宏之; 鈴木智也
電子情報通信学会2008年総合大会, 2008 - 複雑システムにおけるネットワーク中心性が予測精度に与える影響
鈴木智也; 池田真一
数理モデル化と問題解決研究会, 2008 - 時系列データの天底予測のための非線形予測法
鈴木智也; 太田真喜
数理モデル化と問題解決研究会, 2008 - 複雑ネットワークシステムにおけるノードの中心性と予測精度の関係
池田真一; 鈴木智也
電子情報通信学会非線形問題研究会, 2008 - 非線形予測法に基づく時系列データの天底予測
太田真喜; 鈴木智也
電子情報通信学会非線形問題研究会, 2008 - 有向重み付きネットワーク解析
鈴木智也
ネットワークが創発する知能研究会 第3回国内ワークショップ, 2007 - A Novel Clustering Coefficient for Directed and Weighted Networks
鈴木智也
電子情報通信学会2006年総合大会, 2007 - 予測領域推定とリアプノフ指数との関係
原木大典; 鈴木智也; 池口徹
電子情報通信学会総合大会, 2007 - 多次元スパイク列からのニューラルネットワーク構造推定
芦澤徹; 原木大典; 鈴木智也; 池口徹
電子情報通信学会総合大会, 2007 - Estimating Network Structures from Multi-dimensional time series
Tohru Ashizawa; Daisuke Haraki; Tomoya Suzuki; Tohru Ikeguchi
Internal Symposium on Complexity Modeling and its Application, 2006 - Nonlinear Prediction Interval Estimation by the Bootstrap Method
Daisuke Haraki; Tomoya Suzuki; Hiroki Hashiguchi; Tohru Ikeguchi
Internal Symposium on Complexity Modeling and its Application, 2006 - Synchronization in STDP neural network and its network structure
Tohru Ikegchi; Tomoya Suzuki; Ryosuke Hosaka; H. Kato
Internal Symposium on Complexity Modeling and its Application, 2006 - Comparing Predictability with Prediction Error Distribution
原木大典; 鈴木智也; 池口徹
電子情報通信学会2005年総合大会, 2006 - 多変数時系列からのネットワーク構造推定
芦澤徹; 原木大典; 鈴木智也; 池口徹
電子情報通信学会2005年総合大会, 2006 - 結合写像格子で構成された複雑ネットワーク構造の推定
鈴木智也; 池口徹; 堀尾喜彦
電子情報通信学会2005年総合大会, 2006 - Relation between Prediction Accuracy of Nonlinear Modeling and Nonlinearity of Time Series
原木大典; 鈴木智也; 池口徹
電子情報通信学会非線形問題研究会, 2006 - Evaluating Structure of Complex Networks Hidden in Nikkei 225 Stock Market
鈴木智也; 池口徹; 堀尾喜彦
電子情報通信学会非線形問題研究会, 2006 - Estimating Structure of Complex Networks from Time Series
Tomoya Suzuki; Tohru Ikeguchi; Yoshihiko Horio
Internal Symposium on Complexity Modeling and its Application, 2005 - Local Linear Prediction with the Bootstrap Resampling
Daisuke Haraki; Tomoya Suzuki; Tohru Ikeguchi
Internal Symposium on Complexity Modeling and its Application, 2005 - From Random to Small-World Network by STDP Learning
鈴木智也; 池口徹; 鈴木増雄
電子情報通信学会2005年総合大会, 2005 - ブートストラップ法を用いた非線形予測
原木大典; 鈴木智也; 池口徹
電子情報通信学会2005年ソサイエティ大会, 2005 - A Measure for Nonlinear Predictability of Real-Life Chaos
Tomoya Suzuki; Tohru Ikeguchi; Masuo Suzuki
the 8-th Experimental Chaos Conference, 2004 - Deterministic Nonlinearity of Temporal Structures in Internet Traffic Time Series
Tohru Ikeguchi; Hosoda Kento; Tomoya Suzuki; Mikio Hasegawa
the 8-th Experimental Chaos Conference, 2004 - Evaluating the Advantage of Nonlinear Modeling of Multivariable Dynamical System
鈴木智也; 池口徹; 鈴木増雄
電子情報通信学会NLP, 2004 - Estimating Nonlinearity of Time Series and Efficiency of Nonlinear Modeling
鈴木智也; 池口徹; 鈴木増雄
電子情報通信学会2004年ソサイエティ大会, 2004 - A Measure for Nonlinear Predictability on the Basis of Modeling Complexity
鈴木智也; 池口徹; 鈴木増雄
電子情報通信学会2004年総合大会, 2004 - Evaluating Nonlinear Predictability Using Information Criterion and Resampling Method
鈴木智也; 池口徹; 鈴木増雄
電子情報通信学会NLP, 2004 - A Novel Measure for Non-Linear Prediction Accuracy
鈴木智也; 池口徹; 鈴木増雄
電子情報通信学会NLP, 2003 - Dangers of the Method of Surrogate Data for Discretized Time Series
鈴木智也; 池口徹; 鈴木増雄
電子情報通信学会2003年ソサイエティ大会, 2003 - Effects of Data Windows for the Method of Surrogate Data
鈴木智也; 池口徹; 鈴木増雄
電子情報通信学会NLP, 2003 - Surrogate test for chaos game representation
鈴木智也; 池口徹; 鈴木増雄
電子情報通信学会2003年総合大会, 2003 - サロゲートデータ法によるインターネットトラフィックデータの解析
細田健人; 鈴木智也; 長谷川幹雄; 池口徹
電子情報通信学会NLP, 2003 - Dangers of Chaos Game Representation
鈴木智也; 池口徹; 鈴木増雄
電子情報通信学会NLP, 2003 - 相関次元推定における量子化誤差及びノイズの影響
鈴木智也; 池口徹; 鈴木増雄
電子情報通信学会2002年総合大会, 2002 - 量子化された観測データに対するノイズの評価について
鈴木智也; 池口徹; 鈴木増雄
電子情報通信学会NLP, 2002 - 取引時間間隔情報に基づく為替相場の非線形解析
鈴木智也; 池口徹; 鈴木増雄
電子情報通信学会NLP, 2001 - ボラティリィーとスプレッド変化に基づく為替相場の非線形モデル化
鈴木智也; 池口徹; 鈴木増雄
電子情報通信学会NLP, 2001 - 局所線形予測法における近傍探索の一手法について
鈴木智也; 池口徹; 鈴木増雄
平成12年度東京理科大ハイテクリサーチセンター計算科学フロンティアセンター研究報告, 2001 - 局所線形予測法における近傍探索の一手法について
鈴木智也; 池口徹; 鈴木増雄
電子情報通信学会NLP, 2001
Affiliated academic society
Research Themes
- インフラ情報を活用したシ
May 2023 - 有価証券運用における予兆分析への機械学習,AI活用
常陽銀行
Apr. 2022 - Mar. 2023 - Enhancement of Validating Market Hypothesis by Evidence-based Financial Technical Analysis
Grant-in-Aid for Scientific Research (C)
Ibaraki University
Apr. 2020 - Mar. 2023 - ビッグデータを利用した為替市場の予測モデルの構築
Mar. 2019 - Mar. 2023 - AI運用のためのニューステキストからのアルファならびにセンチメント情報自動抽出
Aug. 2018 - Jul. 2022 - 機械学習法を駆使した金融テクニカル分析の科学的妥当性の検証
基盤研究(C)
Apr. 2016 - Mar. 2019 - 背景ダイナミクスを重視する非線形時系列解析と金融工学への応用
Grant-in-Aid for Scientific Research(C)
Jun. 2013 - Mar. 2016 - 動的に変化する多変量複雑システムの動画表現・構造同定・最適化工学への応用
Grant-in-Aid for Young Scientists(B)
Jun. 2010 - Mar. 2012 - ネットワーク構造の推定を基盤とした複雑システムの理解とその応用
Grant-in-Aid for Young Scientists(B)
Jun. 2008 - Mar. 2010 - 進化的計算手法による株価予測モデルの動的最適化
Jun. 2008 - Mar. 2009 - 複雑ネットワーク上で駆動するダイナミクスと発生する非線形非平衡現象の統合的解析
Others
Oct. 2006 - Mar. 2008
Industrial Property Rights
Social Contribution Activities
- いばらきイノベーションアドバイザー
advisor
10 Oct. 2019 - Present - NPO法人 日本テクニカルアナリスト協会 評議員
organizing_member
14 Jun. 2014 - Present - 国際テクニカルアナリスト連盟(IFTA)理事
organizing_member
01 Oct. 2019 - 28 Sep. 2020 - 茨城県産業技術イノベーションセンター機能強化検討委員
advisor
2018 - 2019 - 情報処理学会「数理モデル化と問題解決研究会 (MPS)」運営委員
organizing_member
Apr. 2013 - 2019 - NPO法人 日本テクニカルアナリスト協会(数理研究部)幹事
organizing_member
Apr. 2012 - 2019 - 情報処理学会論文誌「数理モデルと応用 (TOM)」編集委員
editor
Apr. 2006 - 2019
Media Coverage
- AIは料理で有名シェフに勝てるのか?人類VS AIによる料理対決がついに決着!フレンチでは圧勝もイタリアン、オリジナルでAIが逆転
PR TIMES, 26 Mar. 2024, Internet - AIで有価証券運用 常陽銀と茨城大が研究発表
日刊工業新聞, 日刊工業新聞, 24 May 2023, Paper - 投資機会 AIで発見 ソフト実用化へ 常銀と茨大 開発進める
読売新聞, 読売新聞, 24 May 2023, Paper - 株運用などにAIや機械学習 茨城大と常陽銀行が研究報告会
NHK, 17 May 2023, Media report - 常陽銀と茨城大 AI使い株売買判断 共同研究の中間報告会
茨城新聞, 茨城新聞, 17 May 2023, Paper - 常陽銀、有価証券のAI運用探る 茨城大と共同研究が進展
ニッキン, ニッキン, 17 May 2023, Paper - 人間がAIを活用する投資の未来
ONE WALK, FXPEDIA, 03 Mar. 2023, Internet - 今こそ仲値を科学する
FX雑誌「外国為替」vol.3, FX雑誌「外国為替」vol.3, 22 Feb. 2023, Paper - 人間とAI、為替取引においてはどちらが優位になるのか?
有限会社グリーン・アース, 03 Dec. 2022, Internet - 為替取引で人工知能が出来ることは限定的である ー AI詐欺やデマに要注意
World Academic Journal, 23 Nov. 2022, Internet - 為替取引でのAI活用は果たして有効なのか?
エモーショナルリンク, 14 Sep. 2022, Internet - 注目の研究!ゴトオビの仲値トレードが有効な通貨ペア・エントリー時間は?
テクニカルブック, 24 May 2022, Internet - 【神回】FXプロ×茨城大学教授 仲値トレードの神髄に迫る!
ヒロセ通商, 【公式】ヒロセ通商(LION FX)動画チャンネル, 13 Jan. 2022, Internet - AI運用の現状と可能性(下)行動経済学とDXで高める運用技術の透明性
日本金融通信社, ニッキン投信情報, 07 Jun. 2021, Paper - AI運用の現状と可能性(上)持続的に発展するAI技術とビッグデータ
日本金融通信社, ニッキン投信情報, 31 May 2021, Paper - 茨城大、AIで市場心理分析
日本経済新聞, 31 Mar. 2021, Paper - AI+行動経済学で資産運用 大和アセット・茨城大がファンド,「異常」捕捉し銘柄選別
日経ヴェリタス, 28 Mar. 2021, Paper - AI運用の「強み」と「弱み」とは 持ち味は、主観交えず 膨大な情報を瞬時に自動処理 茨城大学大学院 鈴木智也教授に聞く
日本証券新聞, 15 Mar. 2021, Paper - 行動経済学に基づく運用モデル
ニッキン投資情報, 01 Mar. 2021, Paper - 今の株価はコロナバブル? 人工知能で値動き分析すると
朝日新聞, 26 Feb. 2021, Internet - 人間心理の株価への影響 AI検知・茨城大などモデル
日本経済新聞, 24 Feb. 2021, Paper - ビッグデータを解析し株価を予測するAI
ニュートンプレス, Newton 大図鑑シリーズ(AI大図鑑), 15 Dec. 2020, Paper - ビッグデータを解析し株価を予測するAI
Newton, 30 Nov. 2020, Paper - 茨城大学鈴木教授に聞く!ゴトー日の仲値トレードはどうすべき?統計データからわかるアノマリー検証
エフプロ, 20 Nov. 2020, Internet - 茨城大学鈴木教授に聞く!ゴトー日はどういうトレードをするべき?統計データの分析からわかる傾向
エフプロ, 28 Jul. 2020, Internet - AIにより機械化が進む資産運用ビジネス
ニッキン投信情報, 14 Jan. 2020, Paper - 0.1秒後の為替レートを8~9割の精度で予測! AIによる金融市場研究はどこまで進んだ?
ザイFX!, 03 Jul. 2019, Internet - ゴトー日の金曜日の仲値トレードは儲かる! 茨城大・鈴木智也研究室が検証し学会発表
ザイFX!, 01 Jul. 2019, Internet - シンギュラリティーにっぽん
朝日新聞, 朝日新聞, 16 Jun. 2019, Paper - AIを駆使した中古車の価格変動予測が、産学連携の共同研究で実現
プロト総研, 22 Jan. 2019, Internet - 県内ベンチャー紹介
日刊工業新聞, 19 Dec. 2018, Paper - AIで株価予測はどこまでできるのか
Newton, Newton別冊「ゼロからわかる人工知能 (仕事編)」, 17 Dec. 2018, Paper - 産学連携へ茨大工学部と企業の交流会 (AIデータ解析 鈴木智也研究室)
NHK, 27 Nov. 2018, Media report - 茨城大発ベンチャー、鈴木教授がAI関連で起業
朝日新聞, 21 Sep. 2018, Paper - AI活用し株価予想 茨城大教授 ベンチャー設立
茨城新聞, 20 Sep. 2018, Paper - AI活用したベンチャー企業設立
自動車流通新聞, 20 Sep. 2018, Paper - 茨大教授がAIベンチャー 金融や不動産、医療など視野
東京新聞, 20 Sep. 2018, Paper - 茨城大、新ベンチャー設立 AI「集合知」活用サービス提供
日刊工業新聞, 20 Sep. 2018, Paper - 茨城大・鈴木教授がVB設立 AIの集合知で投資判断
日本経済新聞, 19 Sep. 2018, Paper - 茨大研究成果利用ベンチャー設立
NHK, 19 Sep. 2018, Media report - 茨城大教授がベンチャー設立 金融投資にAIを活用
産経新聞, 19 Aug. 2018, Paper - 大和投信,AI運用の研究加速 茨城大学の鈴木教授を招へい
ニッキン投資情報, 12 Mar. 2018, Paper - 投信運用「AI寄れば文殊の知恵」
日経ヴェリタス, 18 Feb. 2018, Paper - 進化する「AI運用」の今
日本証券新聞, 25 Oct. 2017, Paper - 運用各社「AI対応」相次ぐ
R&I格付投資情報センター, ファンド情報, 23 Oct. 2017, Paper - AI運用の開発加速
日経新聞, 13 Oct. 2017, Paper - 年間優秀論文賞の記念講演会「集合知AIは投資効果向上に有効」
J-Money, 28 Jul. 2017, Paper - 株価予想,AIの集合知で
日経新聞, 27 Jul. 2017, Paper - 投資すべき銘柄をAIで選択 鈴木教授の研究成果が最優秀賞に 日本人3人目
毎日新聞, 02 Jun. 2017, Paper - 株価データ分析国際団体賞受賞
日経新聞, 01 Jun. 2017, Paper - AIの「集合知」値動き見抜く
日経ヴェリタス, 16 Apr. 2017, Paper