Forecasting Performance of Nonlinear Models for Intraday Stock Returns

Abstract

We studied the predictability of intraday stock market returns using both linear and nonlinear time series models. For the S&P 500 index we compared simple autoregressive and random walk linear models with a range of nonlinear models, including smooth transition, Markov switching, artificial neural network, nonparametric kernel regression and support vector machine models for horizons of 5, 10, 20, 30 and 60 minutes. The empirical results indicate that nonlinear models outperformed linear models on the basis of both statistical and economic criteria. Specifically, although return serial correlation receded by around 10 minutes, return predictability still persisted for up to 60 minutes according to nonlinear models, even though profitability decreases as time elapses. More flexible nonlinear models such as support vector machines and artificial neural network did not clearly outperform other nonlinear models.

Modelling and Trading the Greek Stock Market with Gene Expression and Genetic Programing Algorithms

“…Other recent literature in the area includes Matias and Reboredo (2012) who forecast the S&P 500 intraday stock market returns with a range of linear (simple autoregressive and random walk models) and non linear models (such as smooth transition, Markov switching, NNs, nonparametric kernel regression and support vector machine) models. Their results indicate the superiority of the non linear models in both statistical and economic terms.…”

Forecasting foreign exchange rates with adaptive neural networks using radial-basis functions and Particle Swarm Optimization

“…al. (2011) won the NN3 Forecasting Competition problem with an innovative approach based on the use of median for combining MLP forecasts and Matias and Reboredo (2012) forecast successfully with NNs and other nonlinear models intraday stock market returns. In a forecasting competition, Dunis et.…”

Multisource Data Analysis for Stock Prediction

“…Technical analysis is a securityanalysis methodology used to forecast the direction of prices by studying past market data-mainly price and volume. The founders of technical analysis believe the stock market is related to the historical data, and the volatility of stock prices can be quantified and forecasted [5]. Technical analysts typically make decisions using theories such as Dow theory and the Elliott wave principle [6].…”

“…In addition, given the general and consistent use of factors by investors, the performance of multifactor models is relatively stable. That said, under different market conditions, various factors can play significant roles [10]. Therefore, investors and quantitative researchers have developed several factor models based on different strategies.…”