PRIMER ON USING NEURAL NETWORKS FOR FORECASTING MARKET VARIABLES

Abstract

Ability to forecast market variables is critical to analysts, economists and investors. Among other uses, neural networks are gaining in popularity in forecasting market variables. They are used in various disciplines and issues to map complex relationships. We present a primer for using neural networks for forecasting market variables in general, and in particular, forecasting volatility of the S&P 500 Index futures prices. We compare volatility forecasts from neural networks with implied volatility from S&P 500 Index futures options using the Barone-Adesi and Whaley (BAW) model for pricing American options on futures. Forecasts from neural networks outperform implied volatility forecasts. Volatility forecasts from neural networks are not found to be significantly different from realized volatility. Implied volatility forecasts are found to be significantly different from realized volatility in two of three cases. A revised version of this paper has since been published in the Journal of Business Research. Please use this version in your citations.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 22 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.