Pub Date : 1996-03-24DOI: 10.1109/CIFER.1996.501826
K. Schierholt, C. Dagli
In recent years, many attempts have been made to predict the behavior of bonds, currencies, stocks, or stock markets. The Standard and Poors 500 Index is modeled using different neural network classification architectures. Most previous experiments used multilayer perceptrons for stock market forecasting. A multilayer perceptron architecture and a probabilistic neural network are used to predict the incline, decline, or steadiness of the index. The results of trading with the advice given by the network is then compared with the maximum possible performance and the performance of the index. Results show that both networks can be trained to perform better than the index, with the probabilistic neural network performing slightly better than the multi layer perceptron.
{"title":"Stock market prediction using different neural network classification architectures","authors":"K. Schierholt, C. Dagli","doi":"10.1109/CIFER.1996.501826","DOIUrl":"https://doi.org/10.1109/CIFER.1996.501826","url":null,"abstract":"In recent years, many attempts have been made to predict the behavior of bonds, currencies, stocks, or stock markets. The Standard and Poors 500 Index is modeled using different neural network classification architectures. Most previous experiments used multilayer perceptrons for stock market forecasting. A multilayer perceptron architecture and a probabilistic neural network are used to predict the incline, decline, or steadiness of the index. The results of trading with the advice given by the network is then compared with the maximum possible performance and the performance of the index. Results show that both networks can be trained to perform better than the index, with the probabilistic neural network performing slightly better than the multi layer perceptron.","PeriodicalId":378565,"journal":{"name":"IEEE/IAFE 1996 Conference on Computational Intelligence for Financial Engineering (CIFEr)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124786102","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1996-03-24DOI: 10.1109/CIFER.1996.501833
M. Makivic
Summary form only given. We propose a combination of the path-integral Monte Carlo method and the maximum entropy method as a comprehensive solution for the problem of pricing of derivative securities. The path-integral Monte Carlo approach relies on the probability distribution of the complete histories of the underlying security, from the present time to the contract expiration date. In our present implementation, the Metropolis algorithm is used to sample the probability distribution of histories (paths) of the underlying security. The advantage of the path integral approach is that complete information about the derivative security, including its parameter sensitivities, is obtained in a single simulation. It is also possible to obtain results for multiple values of parameters in a single simulation. The input to the path-integral Monte Carlo method is the assumed propagator for the stochastic process of the underlying. The path integral method is flexible about the input stochastic process and it can be used for both American and European contracts. Derivative valuation can be viewed as a statistical inference procedure about the underlying stochastic process. In its simplest form it reduces to the computation of implied volatility. It is known that the implied volatility matrix may contain significant variations across strike prices and contract maturities. This implies that parametrization of the underlying process via single volatility parameter is inconsistent with market data. Instead, we formulate an approach which allows one to generate a fully consistent estimate of the complete propagator for the underlying.
{"title":"Path integral Monte Carlo method and maximum entropy: a complete solution for the derivative valuation problem","authors":"M. Makivic","doi":"10.1109/CIFER.1996.501833","DOIUrl":"https://doi.org/10.1109/CIFER.1996.501833","url":null,"abstract":"Summary form only given. We propose a combination of the path-integral Monte Carlo method and the maximum entropy method as a comprehensive solution for the problem of pricing of derivative securities. The path-integral Monte Carlo approach relies on the probability distribution of the complete histories of the underlying security, from the present time to the contract expiration date. In our present implementation, the Metropolis algorithm is used to sample the probability distribution of histories (paths) of the underlying security. The advantage of the path integral approach is that complete information about the derivative security, including its parameter sensitivities, is obtained in a single simulation. It is also possible to obtain results for multiple values of parameters in a single simulation. The input to the path-integral Monte Carlo method is the assumed propagator for the stochastic process of the underlying. The path integral method is flexible about the input stochastic process and it can be used for both American and European contracts. Derivative valuation can be viewed as a statistical inference procedure about the underlying stochastic process. In its simplest form it reduces to the computation of implied volatility. It is known that the implied volatility matrix may contain significant variations across strike prices and contract maturities. This implies that parametrization of the underlying process via single volatility parameter is inconsistent with market data. Instead, we formulate an approach which allows one to generate a fully consistent estimate of the complete propagator for the underlying.","PeriodicalId":378565,"journal":{"name":"IEEE/IAFE 1996 Conference on Computational Intelligence for Financial Engineering (CIFEr)","volume":"310 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122699383","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1996-03-24DOI: 10.1109/CIFER.1996.501844
H. G. Green, R. Martin, M. A. Pearson
An investigation is carried out to demonstrate the effect of data frequency and the use of robust and non-robust techniques for determining risk parameters. The results are for foreign exchange rates but are expected to apply to market price data in general.
{"title":"Robust estimation analytics for financial risk management","authors":"H. G. Green, R. Martin, M. A. Pearson","doi":"10.1109/CIFER.1996.501844","DOIUrl":"https://doi.org/10.1109/CIFER.1996.501844","url":null,"abstract":"An investigation is carried out to demonstrate the effect of data frequency and the use of robust and non-robust techniques for determining risk parameters. The results are for foreign exchange rates but are expected to apply to market price data in general.","PeriodicalId":378565,"journal":{"name":"IEEE/IAFE 1996 Conference on Computational Intelligence for Financial Engineering (CIFEr)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124679075","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1996-03-24DOI: 10.1109/CIFER.1996.501835
O. Castillo, P. Melin
We describe a new method for performing automated mathematical modelling for financial time series prediction using fuzzy logic techniques, dynamical systems and fractal theory. The main idea is that using fuzzy logic techniques we can simulate and automate the reasoning process of human experts in mathematical modelling for financial time series prediction. Our new method for automated modelling consists of three main parts: time series analysis, developing a set of admissible models, and selecting the "best" model. Our method for time series analysis consists of using the fractal dimension of a set of points as a measure of the geometrical complexity of the time series. Our method for developing a set of admissible dynamical systems models is based on the use of fuzzy logic techniques to simulate the decision process of the human experts in modelling financial problems. The selection of the "best" model for financial time series prediction (FTSP) is done using heuristics from the experts and statistical calculations. This new method can be implemented as a computer program and can be considered an intelligent system for automated mathematical modelling for FTSP.
{"title":"Automated mathematical modelling for financial time series prediction using fuzzy logic, dynamical systems and fractal theory","authors":"O. Castillo, P. Melin","doi":"10.1109/CIFER.1996.501835","DOIUrl":"https://doi.org/10.1109/CIFER.1996.501835","url":null,"abstract":"We describe a new method for performing automated mathematical modelling for financial time series prediction using fuzzy logic techniques, dynamical systems and fractal theory. The main idea is that using fuzzy logic techniques we can simulate and automate the reasoning process of human experts in mathematical modelling for financial time series prediction. Our new method for automated modelling consists of three main parts: time series analysis, developing a set of admissible models, and selecting the \"best\" model. Our method for time series analysis consists of using the fractal dimension of a set of points as a measure of the geometrical complexity of the time series. Our method for developing a set of admissible dynamical systems models is based on the use of fuzzy logic techniques to simulate the decision process of the human experts in modelling financial problems. The selection of the \"best\" model for financial time series prediction (FTSP) is done using heuristics from the experts and statistical calculations. This new method can be implemented as a computer program and can be considered an intelligent system for automated mathematical modelling for FTSP.","PeriodicalId":378565,"journal":{"name":"IEEE/IAFE 1996 Conference on Computational Intelligence for Financial Engineering (CIFEr)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121735411","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1996-03-24DOI: 10.1109/CIFER.1996.501816
J. Mulvey
Leading international financial firms are applying multi-stage stochastic programs for managing asset-liability risk over extended time periods. Prominent examples include: Towers Perrin, State Farm Insurance, Falcon Asset Management, Frank Russell and Unilever. The asset-liability management systems assist pension plan investors, banks, insurance companies and other leveraged institutions. Wealthy individuals can benefit by developing careful risk management strategies. The advantages of integrating assets and liabilities are discussed along with a brief comparison of alternative modeling frameworks. We describe the advantages of high-performance computers for solving these difficult nonlinear robust optimization problems.
{"title":"Solving robust optimization models in finance","authors":"J. Mulvey","doi":"10.1109/CIFER.1996.501816","DOIUrl":"https://doi.org/10.1109/CIFER.1996.501816","url":null,"abstract":"Leading international financial firms are applying multi-stage stochastic programs for managing asset-liability risk over extended time periods. Prominent examples include: Towers Perrin, State Farm Insurance, Falcon Asset Management, Frank Russell and Unilever. The asset-liability management systems assist pension plan investors, banks, insurance companies and other leveraged institutions. Wealthy individuals can benefit by developing careful risk management strategies. The advantages of integrating assets and liabilities are discussed along with a brief comparison of alternative modeling frameworks. We describe the advantages of high-performance computers for solving these difficult nonlinear robust optimization problems.","PeriodicalId":378565,"journal":{"name":"IEEE/IAFE 1996 Conference on Computational Intelligence for Financial Engineering (CIFEr)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127635153","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1996-03-24DOI: 10.1109/CIFER.1996.501838
Yiu-ming Cheung, Helen Z. H. Lai, L. Xu
We have recently proposed an architecture called Rival Penalized Competitive Learning and Combined Linear Predictor (RPCL-CLP) to model financial time series with a certain degree of success (Cheung et al., 1995). Experiments have shown that RPCL-CLP outperforms ClusNet (Hsu et al., 1993), but it still has features which can be further improved. We propose a modified version called Adaptive RPCL-CLP which can automatically select the number of the initial cluster nodes for RPCL (Xu et al., 1993) and adaptively train the linear predictor's parameters in each cluster node as well as the gating network. We apply it to the forecasting of foreign exchange rates and the Shanghai stock price. As shown by experiments, this adaptive version is much better than RPCL-CLP, and with a trading system it can bring in more returns in foreign exchange market trading.
我们最近提出了一种名为“对手惩罚竞争学习和组合线性预测器”(RPCL-CLP)的架构，用于对金融时间序列进行建模，并取得了一定程度的成功(Cheung et al.， 1995)。实验表明，RPCL-CLP优于ClusNet (Hsu et al.， 1993)，但仍有可以进一步改进的特点。我们提出了一个改进版本，称为自适应RPCL- clp，它可以自动选择RPCL的初始集群节点数量(Xu et al.， 1993)，并自适应地训练每个集群节点和门控网络中的线性预测器参数。我们将其应用于外汇汇率和上海股票价格的预测。实验表明，该自适应版本比RPCL-CLP要好得多，并且配合交易系统，可以在外汇市场交易中带来更高的收益。
{"title":"Adaptive Rival Penalized Competitive Learning and Combined Linear Predictor with application to financial investment","authors":"Yiu-ming Cheung, Helen Z. H. Lai, L. Xu","doi":"10.1109/CIFER.1996.501838","DOIUrl":"https://doi.org/10.1109/CIFER.1996.501838","url":null,"abstract":"We have recently proposed an architecture called Rival Penalized Competitive Learning and Combined Linear Predictor (RPCL-CLP) to model financial time series with a certain degree of success (Cheung et al., 1995). Experiments have shown that RPCL-CLP outperforms ClusNet (Hsu et al., 1993), but it still has features which can be further improved. We propose a modified version called Adaptive RPCL-CLP which can automatically select the number of the initial cluster nodes for RPCL (Xu et al., 1993) and adaptively train the linear predictor's parameters in each cluster node as well as the gating network. We apply it to the forecasting of foreign exchange rates and the Shanghai stock price. As shown by experiments, this adaptive version is much better than RPCL-CLP, and with a trading system it can bring in more returns in foreign exchange market trading.","PeriodicalId":378565,"journal":{"name":"IEEE/IAFE 1996 Conference on Computational Intelligence for Financial Engineering (CIFEr)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134637806","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1996-03-24DOI: 10.1109/CIFER.1996.501834
J. Molle, F. Zapatero
Very often the dynamics of the interest rate and/or the risk premium do not allow to obtain a close form solution for the price of the pure discount bond. One possible approach is to use Monte Carlo simulation. In order to do this we first have to simulate the path of the stochastic variables. After doing this a number of times, we average over the different realizations. The result will be the price of the bond. In fact, very often it is assumed that the equity risk premium is zero. This is a convenient simplification, but it takes away some of the richness of equilibrium models that assume risk-averse investors.
{"title":"Problems with Monte Carlo simulation in the pricing of contingent claims","authors":"J. Molle, F. Zapatero","doi":"10.1109/CIFER.1996.501834","DOIUrl":"https://doi.org/10.1109/CIFER.1996.501834","url":null,"abstract":"Very often the dynamics of the interest rate and/or the risk premium do not allow to obtain a close form solution for the price of the pure discount bond. One possible approach is to use Monte Carlo simulation. In order to do this we first have to simulate the path of the stochastic variables. After doing this a number of times, we average over the different realizations. The result will be the price of the bond. In fact, very often it is assumed that the equity risk premium is zero. This is a convenient simplification, but it takes away some of the richness of equilibrium models that assume risk-averse investors.","PeriodicalId":378565,"journal":{"name":"IEEE/IAFE 1996 Conference on Computational Intelligence for Financial Engineering (CIFEr)","volume":"103 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116488298","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1996-03-24DOI: 10.1109/CIFER.1996.501824
R. Yager
The problem of selecting an investment option in the face of uncertainty with respect to the payoff is considered. Methods for the representation of uncertainty based on the theory of fuzzy sets and the Dempster-Shafer belief structure are described. Approaches for comparing alternatives under various kinds of uncertainty are discussed.
{"title":"Fuzzy set methods for uncertainty representation in risky financial decisions","authors":"R. Yager","doi":"10.1109/CIFER.1996.501824","DOIUrl":"https://doi.org/10.1109/CIFER.1996.501824","url":null,"abstract":"The problem of selecting an investment option in the face of uncertainty with respect to the payoff is considered. Methods for the representation of uncertainty based on the theory of fuzzy sets and the Dempster-Shafer belief structure are described. Approaches for comparing alternatives under various kinds of uncertainty are discussed.","PeriodicalId":378565,"journal":{"name":"IEEE/IAFE 1996 Conference on Computational Intelligence for Financial Engineering (CIFEr)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128567176","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1996-03-24DOI: 10.1109/CIFER.1996.501847
T. Sutter, G. Mollet, Markus Schröder, R. Kruse, J. Gebhardt
The paper deals with the use of fuzzy set theory in succession planning. In the whole process of management planning, succession planning is one part which plays an important role. Selecting a candidate for a concrete position is particularly difficult if you have a lot of alternatives. The paper proposes a way of flexible querying in ordinary databases. In order to improve query results, the SQL systax is extended to allow a kind of "weak" query with the help of fuzzy methods. This sort of imprecise information retrieval is more convenient for managers who are responsible for personnel recruitment and planning. Queries simplify the decision process by pre-selecting the alternatives. The results of this work are included into a fuzzy based software tool, which is used by the VW AG to select successors for vacant positions.
{"title":"Fuzzy queries for top-management succession planning","authors":"T. Sutter, G. Mollet, Markus Schröder, R. Kruse, J. Gebhardt","doi":"10.1109/CIFER.1996.501847","DOIUrl":"https://doi.org/10.1109/CIFER.1996.501847","url":null,"abstract":"The paper deals with the use of fuzzy set theory in succession planning. In the whole process of management planning, succession planning is one part which plays an important role. Selecting a candidate for a concrete position is particularly difficult if you have a lot of alternatives. The paper proposes a way of flexible querying in ordinary databases. In order to improve query results, the SQL systax is extended to allow a kind of \"weak\" query with the help of fuzzy methods. This sort of imprecise information retrieval is more convenient for managers who are responsible for personnel recruitment and planning. Queries simplify the decision process by pre-selecting the alternatives. The results of this work are included into a fuzzy based software tool, which is used by the VW AG to select successors for vacant positions.","PeriodicalId":378565,"journal":{"name":"IEEE/IAFE 1996 Conference on Computational Intelligence for Financial Engineering (CIFEr)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129653629","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1996-03-24DOI: 10.1109/CIFER.1996.501821
S. Leven
Modelling markets top-down tends to eliminate the dynamic nature of valuation. As prices constitute emergent properties of market forces and these forces emerge from anticipation and interaction of agents, only by employing games based in discursive systems theory can we detect "systems embedded in systems". Human decision-making has long been described as the convolving of habitual, inferential and affective processes. We have designed a series of simulations that employ neural networks to model the biological processes involved in individual and interactive decision-making. We have also designed models employing these interactions in organizational and market processes. Further, we suggest that observer effects are central to the measurement process in time-series analysis, from series and component definition to experimental design through outcome interpretation. Employing a neural network tool called Differential Filtering, we have suggested that these effects can be understood and, to some extent, vitiated. Finally, we have demonstrated the ability of the brain-emulating networks to detect context and to discover texture in data series, as a solution to problems such as data fusion and data decomposition. We discuss these models in light of modern approaches to complex systems information processing.
{"title":"Models of market behavior: bringing realistic games to market","authors":"S. Leven","doi":"10.1109/CIFER.1996.501821","DOIUrl":"https://doi.org/10.1109/CIFER.1996.501821","url":null,"abstract":"Modelling markets top-down tends to eliminate the dynamic nature of valuation. As prices constitute emergent properties of market forces and these forces emerge from anticipation and interaction of agents, only by employing games based in discursive systems theory can we detect \"systems embedded in systems\". Human decision-making has long been described as the convolving of habitual, inferential and affective processes. We have designed a series of simulations that employ neural networks to model the biological processes involved in individual and interactive decision-making. We have also designed models employing these interactions in organizational and market processes. Further, we suggest that observer effects are central to the measurement process in time-series analysis, from series and component definition to experimental design through outcome interpretation. Employing a neural network tool called Differential Filtering, we have suggested that these effects can be understood and, to some extent, vitiated. Finally, we have demonstrated the ability of the brain-emulating networks to detect context and to discover texture in data series, as a solution to problems such as data fusion and data decomposition. We discuss these models in light of modern approaches to complex systems information processing.","PeriodicalId":378565,"journal":{"name":"IEEE/IAFE 1996 Conference on Computational Intelligence for Financial Engineering (CIFEr)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124584244","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：