Further Critique of Garch/Arma/Var/Stochastic-Volatility Models

This article critiques models of market risk (ARMA, GARCH, VAR, Stochastic Volatility, etc.). The author analyzes underlying assumptions of these models and explains how they are incorrect. The existing metrics for quantifying risk such as standard deviation, VAR/GARCH/ARMA/SV, etc., are inadequate; and do not account for many facets of risk and decision making; and do not incorporate the many psychological, legal, liquidity, knowledge, and price-dynamic factors inherent in stock markets and asset prices, and thus, are not useful and accurate in many asset markets, particularly those outside the U.S. and in emerging market countries.Areas for further research include: (a) development of dynamic market-risk models that incorporate asset-market psychology, liquidity, market size, frequency of trading, knowledge differences among market participants, information (capabilities and processing, and trading rules in each market; and (b) further development of concepts in belief systems.