Neural network calibrated stochastic processes: forecasting financial assets

Date

2013-03-01

Author

Giebel, Stefan
Rainer, Martin

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

233
views

0
downloads

If a given dynamical process contains an inherently unpredictable component, it may be modeled as a stochastic process. Typical examples from financial markets are the dynamics of prices (e.g. prices of stocks or commodities) or fundamental rates (exchange rates etc.). The unknown future value of the corresponding stochastic process is usually estimated as the expected value under a suitable measure, which may be determined from distribution of past (historical) values. The predictive power of this estimation is limited by the simplifying assumptions of common calibration methods. Here we propose a novel method of "intelligent" calibration, using learning (2-layer) neural networks in order to dynamically adapt the parameters of a stochastic model to the most recent time series of fixed length (memory depth) to the past. The process parameters are determined by the weights of the intermediate layer of the neural network. The final layer combines these parameters in a meaningful manner yielding the forecasting value for the stochastic process. On each actual finite memory, the neural network is trained by back-propagation, obtaining a much more flexible and realistic parameter calibration than an analogous fit to an autoregressive models could do. In the context of processes related to financial assets, the final combination of the output layer relates to their market-price-of-risk. The back propagation is limited to the typical memory length of the financial market (for example 10 previous business days). We demonstrate the learning efficiency of the new algorithm by tracking the next-day forecasts with one typical examples each, for the asset classes of currencies and stocks.

Subject Keywords

Management Science and Operations Research

URI

https://hdl.handle.net/11511/66032

Journal

CENTRAL EUROPEAN JOURNAL OF OPERATIONS RESEARCH

DOI

https://doi.org/10.1007/s10100-011-0234-3

Collections

Graduate School of Applied Mathematics, Article

Suggestions

OpenMETU
Core

Optimal lot-sizing/vehicle-dispatching policies under stochastic lead times and stepwise fixed costs Alp, O; Erkip, NK; Gullu, R (Institute for Operations Research and the Management Sciences (INFORMS), 2003-01-01) We characterize optimal policies of a dynamic lot-sizing/vehicle-dispatching problem under dynamic deterministic demands and stochastic lead times. An essential feature of the problem is the structure of the ordering cost, where a fixed cost is incurred every time a batch is initiated (or a vehicle is hired) regardless of the portion of the batch (or vehicle) utilized. Moreover, for every unit of demand not satisfied on time, holding and backorder costs are incurred. Under mild assumptions we show that the ...
On the classical Maki-Thompson rumour model in continuous time Belen, Selma; Kropat, Erik; Weber, Gerhard Wilhelm (Springer Science and Business Media LLC, 2011-03-01) In this paper, the Maki-Thompson model is slightly refined in continuous time, and a new general solution is obtained for each dynamics of spreading of a rumour. It is derived an equation for the size of a stochastic rumour process in terms of transitions. We give new lower and upper bounds for the proportion of total ignorants who never learned a rumour and the proportion of total stiflers who either forget the rumour or cease to spread the rumour when the rumour process stops, under general initial condit...
Assessment of criteria - rich rankings for environmental policy making Yeralan, Sencer; Ozdoglar, Mehmet Rasit; Azizoğlu, Meral (Inderscience Publishers, 2011-12-01) This paper illustrates the use of mathematical programming techniques to extract more information out of composite indexes (e.g., the EPI-2008) that would assist decision makers. While recognising the qualitative aspects of such decision making, in order to support and guide the policy making process, we develop analytical tools to assist the process. We carefully delineate our models to be limited only to the provable quantitative properties of the available objective data. However, such data are processed...
Effective optimization with weighted automata on decomposable trees Ravve, E. V.; Volkovich, Z.; Weber, Gerhard Wilhelm (Informa UK Limited, 2014-01-02) In this paper, we consider quantitative optimization problems on decomposable discrete systems. We restrict ourselves to labeled trees as the description of the systems and we use weighted automata on them as our computational model. We introduce a new kind of labeled decomposable trees, sum-like weighted labeled trees, and propose a method, which allows us to reduce the solution of an optimization problem, defined in a fragment of Weighted Monadic Second Order Logic, on such a tree to the solution of effec...
Stochastic differential games for optimal investment problems in a Markov regime-switching jump-diffusion market Savku, E.; Weber, Gerhard Wilhelm (Springer Science and Business Media LLC, 2020-08-01) We apply dynamic programming principle to discuss two optimal investment problems by using zero-sum and nonzero-sum stochastic game approaches in a continuous-time Markov regime-switching environment within the frame work of behavioral finance. We represent different states of an economy and, consequently, investors' floating levels of psychological reactions by aD-state Markov chain. The first application is a zero-sum game between an investor and the market, and the second one formulates a nonzero-sum sto...

Citation Formats

S. Giebel and M. Rainer, “Neural network calibrated stochastic processes: forecasting financial assets,” CENTRAL EUROPEAN JOURNAL OF OPERATIONS RESEARCH, pp. 277–293, 2013, Accessed: 00, 2020. [Online]. Available: https://hdl.handle.net/11511/66032.