Neural network calibrated stochastic processes: forecasting financial assets

Giebel, Stefan; Rainer, Martin

doi:10.1007/s10100-011-0234-3

Cited by 10 publications

(1 citation statement)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Also, the important effect of noise disturbances should be taken into account in studying the dynamics of a financial system by means of the neural network approach. Recently, some stochastic financial neural network models have been reported in [2,10,17].…”

Section: Introductionmentioning

confidence: 99%

An impulsive delay discrete stochastic neural network fractional-order model and applications in finance

Bohner¹,

Stamova²

2018

Filomat

View full text Add to dashboard Cite

In this paper, we propose a new tool for modeling and analysis in finance, introducing an impulsive discrete stochastic neural network (NN) fractional-order model. The main advantages of the proposed approach are: (i) Using NNs which can be trained without the restriction of a model to derive parameters and discover relationships, driven and shaped solely by the nature of the data; (ii) using fractional-order differences, whose nonlocal property makes the fractional calculus a suitable tool for modeling actual financial systems; (iii) using impulsive perturbations, which give an opportunity to control the dynamic behavior of the model; (iv) including a stochastic term, which allows to study the effect of noise disturbances generally existing in financial assets; (v) taking into account the existence of time delayed influences. The modeling approach proposed in this paper can be applied to investigate macroeconomic systems.

show abstract

Section: Introductionmentioning

confidence: 99%