An Optimized Approach to Predict the Stock Market Behavior and Investment Decision Making using Benchmark Algorithms for Naive Investors

The stock market is chaotic nature in general. The market and the trend will be well known only for existing investor and also for market players. Due to chaotic nature it is more difficult for making the investment decision by the new investors'. The literature survey as well the recent research history were failed to emphasize the strong decision making power for the new investor. The Time Series analysis is not only an indicator for the stock index but also for the market behavior. This paper focuses on introducing a new mechanism using the Dynamic Neural Network. The NSE indices Nifty-Midcap50 and Reliance are chosen for analysis. The Nonlinear Autoregressive eXogenous (NARX) Network architecture in association with two bench mark algorithms Levenberg Marquardt (LM) and Scaled Conjugate Gradient (SCG) are used for identifying the market behavior. The outcome of the analysis will provide accurate results for the investment decision making to the naive investors'.