Forecasting stock market movement direction with support vector machine

Stock market movements are affected by numerous factors making it one of the most challenging problems for forecasting. This article attempts to predict the direction of movement of stock and stock indices. The study uses three classifiers - Artificial Neural Network, Random Forest and Support Vector Machine with four different representation of inputs. First representation uses raw data (open, high, low, close and volume), The second uses ten features in the form of technical indicators generated by use of technical analysis. The third and fourth portrayal presents two different ways of converting the indicator data into discrete trend data. Experimental results suggest that for raw data support vector machine provides the best results. For other representations, there is no clear winner regarding models applied, but portrayal of data by the proposed approach gave best overall results for all the models and financial series. Consistency of the results highlight the importance of feature generation and right representation of dataset to machine learning techniques.

Download Full-text

Optimization of Financial Forecasting Based on Least Squares Support Vector Machine

Advanced Materials Research ◽

10.4028/www.scientific.net/amr.760-762.1987 ◽

2013 ◽

Vol 760-762 ◽

pp. 1987-1991

Author(s):

Yun Fa Li

Keyword(s):

Support Vector Machine ◽

Simulated Annealing ◽

Stock Market ◽

Prediction Model ◽

Least Squares ◽

Simulated Annealing Algorithm ◽

Least Square ◽

Support Vector ◽

Stock Investment ◽

Annealing Algorithm

To master the variation regularity of finance, obtain greater benefits in stock investment. study of the support vector machine and application in prediction of stock market. The simulated annealing algorithm to optimize the least squares support vector machine prediction model, and the least square support vector machine and simulated annealing algorithm is described, given the optimal prediction model. Through the research on the simulation of the Hang Seng Index, shows that this method is simple, fast convergence, the algorithm with high accuracy. Has the actual guiding sense for investors, the stock market of the financial firm to operate.

Download Full-text

A Hybrid Least Square Support Vector Machine Model with Parameters Optimization for Stock Forecasting

Mathematical Problems in Engineering ◽

10.1155/2015/231394 ◽

2015 ◽

Vol 2015 ◽

pp. 1-7 ◽

Cited By ~ 10

Author(s):

Jian Chai ◽

Jiangze Du ◽

Kin Keung Lai ◽

Yan Pui Lee

Keyword(s):

Support Vector Machine ◽

Least Squares ◽

Support Vector Machine Model ◽

Optimization Methods ◽

Parameters Optimization ◽

Least Square ◽

Movement Direction ◽

Support Vector ◽

Machine Model ◽

Mode Decomposition

This paper proposes an EMD-LSSVM (empirical mode decomposition least squares support vector machine) model to analyze the CSI 300 index. A WD-LSSVM (wavelet denoising least squares support machine) is also proposed as a benchmark to compare with the performance of EMD-LSSVM. Since parameters selection is vital to the performance of the model, different optimization methods are used, including simplex, GS (grid search), PSO (particle swarm optimization), and GA (genetic algorithm). Experimental results show that the EMD-LSSVM model with GS algorithm outperforms other methods in predicting stock market movement direction.

Download Full-text