The Effect of Behavioral Factors on Stock Price Prediction using Generalized Regression and Backpropagation Neural Networks Models

The Effect of Behavioral Factors on Stock Price Prediction using Generalized Regression and Backpropagation Neural Networks Models

Payam Hanafizadeh (Department of Industrial Management, Allameh Tabataba'i University, Tehran, Iran) and Ahmad Hashemi (University of Science and Culture, Tehran, Iran)
Copyright: © 2014 |Pages: 14
DOI: 10.4018/IJBIR.2014100104
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

With regard to the importance of behavioral factors on stock price, which has been mentioned by researchers, this study includes four behavioral factors (overconfidence, representativeness, over reaction and under reaction) in addition to fundamental and technical factors as inputs for neural network models to evaluate the effectiveness of these behavioral factors on stock price prediction accuracy of 10 companies of DJIA index. Multi-layer perceptron (MlP) and generalized regression neural networks are used in this research as models to find the best model for each company based on unique characteristics of its own financial data. This study shows the mentioned behavioral factors are effective on accuracy of predictions of 8 out of 10 companies.
Article Preview

2. Literature Review

As mentioned above behavioral finance tries to explain how unreasonable reactions can be influential in markets. Shefrin (2000) organized psychological phenomena pervading the landscape of finance into three themes: 1) heuristic-driven bias; 2) frame dependence; and 3) inefficient markets. A heuristic is the process by which people seek things out for themselves, usually by trial and error; while trial and error can develop into general rules of thumb, it can also culminate in further errors (Shefrin, 2000). In financial markets, these biases include representativeness, overreaction, underreaction and overconfidence, as well as availability bias, anchoring-and adjustment, aversion to ambiguity and gambler's fallacy, and stock market prediction. The second theme of behavioral finance presents, frame dependence, distinguishes between form and substance in decision-making. It reflects a mix of cognitive and emotional elements. The third one is about common question in finance, whether the market is efficient or inefficient (Lee, 2009). This study focuses on four biases: representativeness, overreaction, underreaction and overconfidence. We want to evaluate the impact of thesis biases on prediction accuracy.

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 8: 2 Issues (2017)
Volume 7: 2 Issues (2016)
Volume 6: 2 Issues (2015)
Volume 5: 4 Issues (2014)
Volume 4: 4 Issues (2013)
Volume 3: 4 Issues (2012)
Volume 2: 4 Issues (2011)
Volume 1: 4 Issues (2010)
View Complete Journal Contents Listing