A goal of the financial sector and of its stakeholders is to estimate the volatility of the stock market. This article proposes a modern approach to solve the problem by means an innovative analysis. The focus of analysis is to improve the estimation of a market index carried through a ARIMA model and to estimate the level of influence of every independent variables (commodities and other principal financial variables) on a dependent variables (market index). Also we are interested in how the set of independent variables is related to the dependent variables. This purpose has required the development of an ad hoc analysis procedure in which we implemented two multivariate algorithms: Artificial Neural Network and Random Forest. The application of two different techniques can evaluate a possible inaccuracy in the model estimation.
Machine learning techniques for stock market prediction.
Najada Firza
;Alfonso Monaco
2015-01-01
Abstract
A goal of the financial sector and of its stakeholders is to estimate the volatility of the stock market. This article proposes a modern approach to solve the problem by means an innovative analysis. The focus of analysis is to improve the estimation of a market index carried through a ARIMA model and to estimate the level of influence of every independent variables (commodities and other principal financial variables) on a dependent variables (market index). Also we are interested in how the set of independent variables is related to the dependent variables. This purpose has required the development of an ad hoc analysis procedure in which we implemented two multivariate algorithms: Artificial Neural Network and Random Forest. The application of two different techniques can evaluate a possible inaccuracy in the model estimation.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
