International Journal of Computer Applications (0975 – 8887) Volume 76– No.17, August 2013 12 A Comparative Study on FFNN and ARIMA Model in the Presence of Outliers K.Senthamarai kannan Professor Department of Statistics Manonmaniam Sundaranar University V.Deneshkumar Research Scholar Department of Statistics Manonmaniam Sundaranar University S.Arumugam CEO Nandha Educational Institutions ABSTRACT Time series data mining (TSDM) techniques explores large amount of time series data in search of interesting relationships among variables. The TSDM methods overcome limitations including stationarity and linearity requirements of traditional time series analysis by adapting data mining concepts for analyzing time series data. The Feed Forward Neural Net is one of the most widely used neural nets. In this paper, the Feed Forward Neural Nets architecture is examined and compared with Statistical Time Series Auto regressive integrated moving average (ARIMA) model for prediction of agricultural production. The performance by ANN model and Time series model for prediction are examined using visualization technique and statistical test and the results are illustrated numerically and graphically Keywords Feed Forward Neural Nets, ANN, ARIMA, Outliers, Forecasting and MSE. 1. INTRODUCTION Artificial neural networks (ANNs) have attracted increasing attentions in recent years for solving many real-world problems. Neural network computing is a key component of any data mining. Neural networks have been used in many business applications for pattern recognition, forecasting, prediction, and classification. ANN is an information processing paradigm that is inspired by the way of biological nervous systems, such as the brain, process information. The key element of this paradigm is the novel structure of the information processing system. It is composed of a large number of highly interconnected processing elements working in unison to solve specific problems. Learning in biological systems involves adjustments to the synaptic connections that exist between the neurons. As the neural network possesses segmentation and identification ability (Zhang et al., 1998) it is widely used in various aspects. For example it is helpful for credit card fraud, stock prices, interest rates and bankruptcy prediction, financial analysis, weather forecasting, judgment for medical images and fingerprint recognition system (Fish et al., (1995), Lee and Chiu, (2002)). Time series is a sequence of observations arranged in specific order of time. The main goals of time series analysis are identifying the nature of the phenomenon represented by the sequence of observations and Forecasting (predicting future values of the time series variable). Arrangement of statistical data in chronological order is known as “Time Series”. Such series have a unique important place in the field of Economic and Business statistics. An economist is interested in estimating the likely population in the coming year so that proper planning can be carried out with regard to food supply, job for the people etc. Similarly, a business man is interested in finding out his likely sales in the near future, so that the businessman could adjust his production accordingly and avoid the possibility of inadequate production to meet the demand. In this connection one usually deal with statistical data, which are collected, observed or recorded at successive intervals of time. Such data are generally referred as „time series‟ data. The list of areas in which time series is observed and studied is endless. Outlier detection is important in many fields. In statistics, an outlier is a observation that is numerically far away from the rest of the data. The handling of outlying observations in a data set is one of the most important tasks in data pre- processing. For many data mining applications, finding the outliers is more interesting than finding the common patterns of the data. Data that have been incorrectly entered or that do not belong to the population from which the rest of the data came can bias estimates and give misleading results. Methods have been devised to identify outlier observations in a variety of situations. With recent advances in technology, scientists are collecting larger data sets and the analyst is getting further and further from the data or even sees every data point. So, it is important to have good methodology for dealing with outlier observations that might not be noticed in a typical data analysis. Artificial neural networks (ANN) have been developed as generalizations of mathematical models of biological nervous systems. A first wave of interest in neural networks (also known as connectionist models or parallel distributed processing) emerged after the introduction of simplified neurons by McCulloch and Pitts (1943). A typical artificial neuron and the modelling of a multilayered neural network are illustrated in Figure 1 and Figure 2 the signal flow from inputs n 2 1 x , .... x , x is considered to be unidirectional, which are indicated by arrows, as is a neuron‟s output signal flow (O). The neuron output signal O is given by the following relationship               n 1 j j j x f net f o ... (1)