Major crises in many of the international financial institutions in the 1990s, due to adverse market moves and poor internal management, raised many questions about the practice of risk management. The need for a foolproof system for measuring risk was felt by practitioners of financial profession. The concept of Value-at-Risk (VaR) was the academicians' answer to this challenge. This study analyzes the performance of VaR techniques by subjecting their prediction to elaborate back testing. The analysis was carried out in the background of Indian capital market, using the Nifty and the Nifty Junior daily returns as the data. The study used Garch and Tgarch models and found that the TgarcH model performed better than the Garch model in predicting VaR. Further, a Vector Autoregression (VAR) framework applied to the two indices showed that Nifty, the broader of the two markets, led in the case of generation of risk in the system.

Certainty is important for smooth transaction of business. But every business normally is risky, i.e., uncertain. It is especially so in financial markets. There are mainly four types of risks in financial markets, namely, credit risk, operational risk, liquidity risk, and market risk. Credit risk arises when the counterparty in an agreement fails to honor its commitment and the loss is associated with that breach of contract. Operational risk involves the errors in making payments or settling transactions, and includes the risk of fraud and regulatory risks. Liquidity risk is caused by an unexpected large and stressful negative cash flow over a very short period. Finally, market risk is the loss to an investment portfolio due to the adverse changes in the price of the financial assets and liabilities, or it is a risk caused solely by market conditions.

With the increasing activity in the financial market, specifically the stock markets, volatility and therefore the market risk of exposure have also grown to be sizable. One method of assessing the market risk of a portfolio is through Value-at-Risk (VaR). It is the maximum expected loss that a portfolio can incur over a certain period of time with a particular level of confidence.

Applied Finance Journal, Risk Management, Vector Autoregression, Financial Markets, Operational Risk, Credit Risk, Generalized Error Distribution, Asian Financial Crisis, Monte Carlo Simulation, GARCH Model, Empirical Analysis, TGARCH Model, Stock Markets, Economic Policies, Emerging Markets.