Indranil Ghosh

Forecasting Volatility in Indian Stock Market using Artificial Neural Network with Multiple Inputs and Outputs

Volatility in stock markets has been extensively studied in the applied finance literature. In this paper, Artificial Neural Network models based on various back propagation algorithms have been constructed to predict volatility in the Indian stock market through volatility of NIFTY returns and volatility of gold returns. This model considers India VIX, CBOE VIX, volatility of crude oil returns (CRUDESDR), volatility of DJIA returns (DJIASDR), volatility of DAX returns (DAXSDR), volatility of Hang Seng returns (HANGSDR) and volatility of Nikkei returns (NIKKEISDR) as predictor variables. Three sets of experiments have been performed over three time periods to judge the effectiveness of the approach. General Terms Artificial Neural Network, Volatility

Using Clustering Method to Understand Indian Stock Market Volatility

In this paper we use Clustering Method to understand whether stock market volatility can be predicted at all, and if so, when it can be predicted. The exercise has been performed for the Indian stock market on daily data for two years. For our analysis we map number of clusters against number of variables. We then test for efficiency of clustering. Our contention is that, given a fixed number of variables, one of them being historic volatility of NIFTY returns, if increase in the number of clusters improves clustering efficiency, then volatility cannot be predicted. Volatility then becomes random as, for a given time period, it gets classified in various clusters. On the other hand, if efficiency falls with increase in the number of clusters, then volatility can be predicted as there is some homogeneity in the data. If we fix the number of clusters and then increase the number of variables, this should have some impact on clustering efficiency. Indeed if we can hit upon, in a sense, an optimum number of variables, then if the number of clusters is reasonably small, we can use these variables to predict volatility. The variables that we consider for our study are volatility of NIFTY returns, volatility of gold returns, India VIX, CBOE VIX, volatility of crude oil returns, volatility of DJIA returns, volatility of DAX returns, volatility of Hang Seng returns and volatility of Nikkei returns. We use three clustering algorithms namely Kernel K-Means, Self Organizing Maps and Mixture of Gaussian models and two internal clustering validity measures, Silhouette Index and Dunn Index, to assess the quality of generated clusters.

Understanding and Forecasting Stock Market Volatility Through Wavelet Decomposition, Statistical Learning and Econometric Methods

Volatility in stock markets evokes varying responses from market participants. While some perceive it as opportunity to make money, others perceive it as a threat and start unwinding their positions. In today’s globalized environment, increased volatility reflects not only the domestic macroeconomic state, but also global uncertainty. While volatility in the stock market as a whole can be influenced by events like oil price shocks, increase in rates of interest in the US and domestic elections, volatility in individual stock prices can be due to perceived growth prospects of the company or the sector, company specific news or policy announcements that can affect a company/sector. In this study, associations, causal influence among three volatility indicators namely, CBOE VIX, INDIA VIX and Historic Volatility (HV) have been carefully examined, and predictive models for forecasting have been developed. An integrated framework incorporating Wavelet decomposition, statistical predictive modeling and standard econometric methods is presented to accomplish the research objectives.

Predicting Stock Returns of Mid Cap Firms in India – An Application of Random Forest and Dynamic Evolving Neural Fuzzy Inference System

The paper examines the pattern of stock returns of mid cap Indian companies over a period of time and proposes frameworks for predictive modelling. Ten features are identified as predictors of stock returns. Subsequently two Machine Learning models, Random Forest and Dynamic Neural Fuzzy Inference System have been employed to check whether the returns can be predicted or not. Experimental setups have been designed and predictive accuracy of the respective models are evaluated using some standard measures. Further, investigation has also been made to recognize the key influential predictors by assessing their impact applying Genetic Algorithm. Our findings suggest that the returns of stocks of mid cap organizations in India can efficiently be forecasted using the frameworks discussed.

Fractal Investigation and Maximal Overlap Discrete Wavelet Transformation (MODWT)-based Machine Learning Framework for Forecasting Exchange Rates

Abstract Foreign currency is bought and sold in the financial markets, every minute, every day, on trading days, like any commodity or stocks of companies. The players in this market are (a) people with underlying interest in foreign currency such as exporters and importers who are continuously hedging in futures or options markets, (b) speculators and (c) arbitrageurs. This paper focuses on this microeconomic flavour of foreign currency as a continuously tradable product and presents a granular framework for forecasting the exchange rate. We initially investigate year-wise inherent nature of movements of three exchange rates, namely Indian rupee/US dollar, Indian rupee/euro and Indian rupee/Japanese yen, during 2011–2016 through Mandelbrot’s single fractal model. Subsequently, maximal overlap discrete wavelet transformation (MODWT) is used to decompose the time series of the individual exchange rates. Random forest and bagging are applied on the decomposed components for predictive modelling.

Application of Multi-Criteria Decision Making Models in Regulatory Evaluation of Commercial Banks in India and its Consistency with Public Perception

This paper, first, examines the performance of a sample set of public sector banks (nationalized banks and State Bank of India) and a sample set of private sector banks in terms of certain regulatory variables. It uses aggregation techniques like TOPSIS, VIKOR and ELECTRE - III to rank their performance. This exercise is done over a period of time to see whether the ranking of banks, both in the private sector and the public sector, has undergone any change. These aggregation techniques provide an alternative method to that of CAMELS used by Reserve Bank of India. Second, the paper delves into market perception of these two sets of banks and ranks them in terms of the aggregation techniques. The paper then combines the above two approaches to check whether the market at large understands the regulatory framework that the banks need to follow. Variables like RISK (Risk Weighted Assets/Total Assets), CAR (Actual Capital Adequacy Ratio), CRRISK (Provisions/Total Advances), VULNER (Ratio of Deposits to Risk Weighted Assets), CLSTATE (Ratio of Government Security Holdings to Total Assets), ROA (Returns on Assets), Net Interest Margin and NPA percentage are selected to represent regulatory variables. Price/Earnings (P/E) Ratio, Price/Book Value per Share (P/BVPS), Size as measured by size of deposits, Dividend Payout Ratio, Dividend Yield and ROE are chosen to represent public perception banks.