Modeling intraday information in financial markets with the scatter search metaheuristic
- 1 June 2015
- journal article
- research article
- Published by World Scientific Pub Co Pte Ltd in International Journal of Financial Engineering
- Vol. 02 (02)
Abstract
Intraday information about stocks or financial indexes is important for the definition of investors' strategies. In this paper two problems where the intraday information is used are studied: (i) modeling a bandwidth for the range of daily maximum and daily prices/values; (ii) modeling an upper and a lower bounds for the daily maximum and daily prices/values. A non-linear transformation of a modified AR(p) process is used, with the parameters computed by the scatter search metaheuristic. The approach is tested with historical data from NASDAQ, DAX, CAC40 and S&P financial series.Keywords
This publication has 23 references indexed in Scilit:
- A high–low model of daily stock price rangesJournal of Forecasting, 2008
- Time series forecasting with a non-linear model and the scatter search meta-heuristicInformation Sciences, 2008
- An empirical model of daily highs and lowsInternational Journal of Finance & Economics, 2007
- Time-series forecasting using flexible neural tree modelInformation Sciences, 2005
- Time series forecasting using a hybrid ARIMA and neural network modelNeurocomputing, 2003
- Time-series forecasting using GA-tuned radial basis functionsInformation Sciences, 2001
- Drift Independent Volatility Estimation Based on High, Low, Open, and Close PricesThe Journal of Business, 2000
- Non-linear financial time series forecasting - Application to the Bel 20 stock market indexEuropean Journal of Economic and Social Systems, 2000
- Time series forecasting by combining RBF networks, certainty factors, and the Box-Jenkins modelNeurocomputing, 1996
- Using neural networks to forecast the S&P 100 implied volatilityNeurocomputing, 1996