M. Thenmozhi

Forecasting Stock Index Movement: A Comparison of Support Vector Machines and Random Forest

There exists vast research articles which predict the stock market as well pricing of stock index financial instruments but most of the proposed models focus on the accurate forecasting of the levels (i.e. value) of the underlying stock index. There is a lack of studies examining the predictability of the direction/sign of stock index movement. Given the notion that a prediction with little forecast error does not necessarily translate into capital gain, this study is an attempt to predict the direction of S&P CNX NIFTY Market Index of the National Stock Exchange, one of the fastest growing financial exchanges in developing Asian countries. Random forest and Support Vector Machines (SVM) are very specific type of machine learning method, and are promising tools for the prediction of financial time series. The tested classification models, which predict direction, include linear discriminant analysis, logit, artificial neural network, random forest and SVM. Empirical experimentation suggests that the SVM outperforms the other classification methods in terms of predicting the direction of the stock market movement and random forest method outperforms neural network, discriminant analysis and logit model used in this study.

Support Vector Machines Approach to Predict the S&P CNX NIFTY Index Returns

The present study, investigates the predictability of S&P CNX NIFTY Index returns using Support vector machines (SVM). The performance of the SVM model in forecasting Nifty index returns is rigorously evaluated in terms of widely used statistical metrics like mean absolute error, root mean square error, normalized mean square error, correctness of sign and direction change (Pesaran and Timmermann (1992, DA test), and equal forecast accuracy using Diebold and Mariano (1995, DM test) by comparing its performance with those of neural network, random forest regression and a linear ARIMA model. The four competiting models are also examined in terms of various trading performance and economic criteria like annualized return, Sharpe ratio, maximum drawdown, annualized volatility, average gain/loss ratio etc via a trading experiment. The findings of the study reveal that SVM model achieves greater forecasting accuracy and improves prediction quality compared to other models experimented in the study. The SVM model can be used as an alternative forecasting tool for Nifty Index returns and it will lead to better returns based on the traditional forecasting accuracy measures, such as root mean squared errors, and financial criteria, such as risk-adjusted measures of return.

Forecasting stock returns based on information transmission across global markets using support vector machines

Abstract This paper provides evidence that forecasts based on global stock returns transmission yield better returns in day trading, for both developed and emerging stock markets. The study investigates the performance of global stock market price transmission information in forecasting stock prices using support vector regression for six global markets—USA (Dow Jones, S&P500), UK (FTSE-100), India (NSE), Singapore (SGX), Hong Kong (Hang Seng) and China (Shanghai Stock Exchange) over the period 1999–2011. The empirical analysis shows that models with other global market price information outperform forecast models based merely on auto-regressive past lags and technical indicators. Shanghai stock index movement was predicted best by Hang Seng Index opening price (57.69), Hang Seng Index by previous day’s S&P500 closing price (54.34), FTSE by previous day’s S&P500 closing price (57.94), Straits Times Index by previous day’s Dow Jones closing price (54.44), Nifty by HSI opening price (60), S&P500 by STI closing price (55.31) and DJIA by HSI opening price (55.22), and Nifty was found to be the most predictable stock index. Trading using global cues-based forecast model generates greater returns than other models in all the markets. The study provides evidence that stock markets across the globe are integrated and the information on price transmission across markets, including emerging markets, can induce better returns in day trading.

Do cross-border acquisitions involving emerging market firms create value

Purpose – The purpose of this paper is to contribute to M&A literature by explicitly investigating whether cross-border acquisitions involving emerging markets, either as acquirers or as targets, create value and how is the performance outcome in such acquisitions impacted by deal-specific characteristics. Design/methodology/approach – This study uses industry-adjusted operating performance to measure acquisition gains, the Wilcoxon signed rank test to examine value creation potential and OLS regression to evaluate the impact of deal characteristics on acquisition gains. Findings – The authors find very pronounced value destruction when emerging market firms acquire targets in developed markets, the adverse outcome being further aggravated when the mode of acquisition is “tender offer” rather than a “negotiated deal”. On the other hand, when developed market firms acquire targets from emerging markets, there is an even chance of value creation, the outcome being favourably influenced by the pre-acquisitio...

Forecasting stock index returns using ARIMA-SVM, ARIMA-ANN, and ARIMA-random forest hybrid models

The purpose of this paper is to develop and identify the best hybrid model to predict stock index returns. We develop three different hybrid models combining linear ARIMA and non-linear models such as support vector machines (SVM), artificial neural network (ANN) and random forest (RF) models to predict the stock index returns. The performance of ARIMA-SVM, ARIMA-ANN and ARIMA-RF are compared with performance of ARIMA, SVM, ANN and RF models. The various competing models are evaluated in terms of statistical metrics and trading performance criteria via a trading strategy. The analysis shows that the hybrid ARIMA-SVM model is the best forecasting model to achieve high forecast accuracy and better returns.

Causal effect of volume on stock returns and conditional volatility in developed and emerging market

This study examines the dynamic causal (linear as well as non-linear) relationship between trading volume and return and between volatility and returns. We have used the vector autoregression based Granger causality framework to examine the linear causality, while the non-linear causality have been investigated using bivariate noisy Macke-Glass model which has been used so far in only economic and commodity data sets. The results of linear and non-linear causality show that trading volume Granger does not cause returns and volatility and suggests that there is unidirectional causality from returns to volume and from volatility to volume. The results strongly support the noise trader model, partially support the sequential information model hypothesis, and contradict the efficient market hypothesis. The evidence that volume does not influence stock returns and volatility can be incorporated by market participants in their trading strategies.

Evidence on changes in time varying volatility around bonus and rights issue announcements

The short term and long term stock price volatility changes around bonus and rights issue announcements have been examined using historical volatility estimation and time varying volatility approach. The results show that the historical volatility has increased after bonus and rights issue announcements. Volatility persistence and unconditional volatility have also increased after the bonus and rights issue announcements. The results support the finding of Medeiros and Matsumoto (2006) but are contrary to the results of Li and Engle (1998), Connoly and Stivers (2005), and Boyd et al. (2005), who report decrease in volatility following the event announcements. This evidence, extendable to any other type of issue announcement, is consistent with theories stating that volatility increases after the seasoned capital issue announcements.

Impact of perceived organisational support on job performance among healthcare professionals

The study examined the impact of perceived organisational support on in-role performance and extra performance towards supervisors, co-workers and patients. A cross-sectional survey was conducted among medical officers and staff nurses working in primary health centres of Tamilnadu. From the results, it is found that perceived organisational support showed a positive impact of on in-role performance and extra role performance. It is found the extra role performance towards patients is greater compared to other job performance measures. The results of the study emphasise the public health department has to identify and reward substantial performers and review welfare policies that are perceived to be unattractive by healthcare professionals.

A hybrid ARIMA-EGARCH and Artificial Neural Network model in stock market forecasting: evidence for India and the USA

This study develops a hybrid model that combines Autoregressive Integrated Moving Average (ARIMA), Exponential GARCH (EGARCH) and Artificial Neural Network (ANN) to predict the daily returns of S&P CNX Nifty and S&P 500 indices by modifying Zhang’s (2003) approach. The performance of the hybrid ARIMA-EGARCH-ANN model is benchmarked against the ARIMA-EGARCH and ANN models. The empirical evidence provides superiority of the hybrid ARIMA-EGARCH-ANN model in terms of the traditional forecasting accuracy measures and Sign and directional change and delivers consistent results for the two time series. This endorses hybrid model robustness and provides its practical use in formulating a strategy for trading in the S&P 500 and Nifty indices.

Behavioural Asset Pricing: Review and Synthesis:

This article presents an overview of literature on behavioural and experimental asset pricing theory. We systematically review the evolution and current development of behavioural asset pricing models as an alternate approach to asset pricing in financial economics literature. A review and synthesis of research carried out in behavioural finance spreading across theoretical, empirical and experimental approaches are presented to understand the behavioural dimension of pricing of financial assets. From theoretical perspective, behavioural asset pricing models try to adopt additional behavioural variables into asset pricing process. In terms of empirical investigation perspective, it is documented that econometric and computational advancement takes its biggest place ever in financial literature when compared with the other field. Our review underlines the fact that the direction of advancing a methodology is changing from financial literature to economics due to the fact that there is huge account of raw data available to analyze. Future research direction should be judging the empirical power of the asset pricing models and their role in practice for incorporating a new dimension to the model. The distinctiveness of the study is that this is the first attempt to review literature written on behavioural asset pricing models in the form of structural empirical review. In doing so, the historical perspective of the concept and the place it will take in future are clarified and the way further researches will be conducted are explored. JEL: E03, G02, G12