Stock Market Regime Detection

Under the direction of Dr. Giancarlo Schrementi The stock market frequently undergoes behavior modification due to a variety of factors such as changes in economic conditions, monetary policy, government policy, and investor sentiment. Such behavior modifications can be categorized by time periods called market regimes. It is important to detect regime changes to optimize quantitative trading and investment strategies. This research paper uses a Hidden Markov Model (HMM) to identify three main market regimes: bull, bear, and neutral, for the S&P 500 Index. The model infers the underlying regime state based on the visible asset returns data. Using the fact that companies in the index are categorized by the Global Industry Classification Standard (GICS), this paper compares the performance of the information technology and energy sectors in the varying regimes