Technical Analysis of Financial Time Series Data & Forecasting Using ARMA and ANN

Abstract

Technical analysis uses past price movements to predict future price movements. It focuses on the market prices themselves, rather than other factors that might affect them. Technical analysis is a method of evaluating securities by statistically analysing their historical trading data. Price pattern can be used as a trading rule to buy and to sell based on price and volume transformation. Not only pattern of data but also important is the future value of stock, here we first consider pattern analysis part and later we observe actual valuation. We have discussed several technical indicators, such as Moving Average (Simple and Exponential), Typical Price, RSI, PVO, Bollinger bands and a new generalized band. These indicators are useful for short term investors to see buying and selling patterns. Depending upon the nature of the series of data, various forecasting models are available in the literature, here we apply Autoregressive Moving Average (ARMA) and Artificial Neural Network (ANN) on agriculture data. ARMA is useful for prediction of future value for linear stationary time series while ANN is useful for predicting future price for non linear time series data. The percentage fit and mean square error (MSE) occurred by ARMA and ANN are finally compared.