Elsevier

Neurocomputing

Volume 10, Issue 2, March 1996, Pages 183-195
Neurocomputing

Paper
Using neural networks to forecast the S&P 100 implied volatility

https://doi.org/10.1016/0925-2312(95)00019-4Get rights and content

Abstract

The implied volatility, calculated using the Black-Scholes model, is currently the most popular method of estimating volatility and is considered by traders to be a significant factor in signalling price movements in the underlying market. Thus, the ability to develop accurate forecasts of future volatility allows a trader to establish the proper strategic position in anticipation of changes in market trends. A neural network has been developed to forecast future volatility, using past volatilities and other options market factors. The performance of this network demonstrates its value as a predictive tool.

References (21)

There are more references available in the full text version of this article.

Cited by (59)

  • Online adaptive machine learning based algorithm for implied volatility surface modeling

    2019, Knowledge-Based Systems
    Citation Excerpt :

    A few examples are as follows. Malliaris and Salchenberger [18] apply artificial neural nets to forecast S&P100 implied volatility with past volatilities and other options market factors. Fengler et al. [19] model IVS dynamics using a semiparametric factor model by means of a principal component analysis, with empirical experiments using the DAX index options data.

  • Machine learning versus econometric jump models in predictability and domain adaptability of index options

    2019, Physica A: Statistical Mechanics and its Applications
    Citation Excerpt :

    Most machine learning methods forecasted the prices of financial derivatives with the expectation that the process of underlying assets will be represented implicitly as a learning function of input variables without the explicit form of return processes. Successful machine learning models for predicting financial derivatives include artificial neural networks (NNs) [1–7], support vector machines [8,9], and Gaussian processes (GPs) [10–13]. These models have also considered different types of available market information, but did not consider explicit formulation for underlying processes.

View all citing articles on Scopus
View full text