ISSN (Print): 2328-3491, ISSN (Online): 2328-3580, ISSN (CD-ROM): 2328-3629 American International Journal of Research in Science, Technology, Engineering & Mathematics AIJRSTEM 14-133; © 2014, AIJRSTEM All Rights Reserved Page 60 AIJRSTEM is a refereed, indexed, peer-reviewed, multidisciplinary and open access journal published by International Association of Scientific Innovation and Research (IASIR), USA (An Association Unifying the Sciences, Engineering, and Applied Research) Available online at http://www.iasir.net Application of Structural Time Series model for forecasting Gram production in India D.P. Singh* 1 A.K. Thakur 2 and D.S. Ram 3 Shaheed Gundadhoor College of Agriculture and Research Station (Indira Gandhi Krishi Vishwavidyalaya), Kumhrawand, Jagdalpur, Bastar (C. G.) 494 005, INDIA. I. INTRODUCTION ARIMA time series methodology is widely used for modelling time-series data. This methodology can be applied only when either the series under consideration is stationary or it can be made so by differencing, de- trending, or by any other means. Another disadvantage is that this approach is empirical in nature and does not provide any insight into the underlying mechanism. An alternative mechanistic approach, which is quite promising, is the Structural time series modelling (Harvey, 1996). Here, the basic philosophy is that characteristics of the data dictate the particular type of model to be adopted from the family. Purpose of present paper is to discuss STM methodology utilized for modelling time-series data in the present of trend, seasonal and cyclic fluctuations. Structural time series model are formulated in such a way that their components are stochastic, i.e. they are regard as being driven by random disturbances. Forecasts are made by extrapolating these components into the future. Harvey and Todd (1983) compare the forecast made by a basic form of the structural model with the forecast made by ARIMA models and conclude that there may be strong arguments in favour of using structural models in practice. Structural models are applicable in the same situations where Box- Jenkins ARIMA models are applicable; however, the structural models tend to be more informative about the underlying stochastic structure of the series. In another paper Harvey (1985) show structural models can be used to model cycle in macro economics time series. Other studied included Kitagawa and Gersch (1984). The forecast obtained from particular model depend on certain variance parameter. The key to handling structural time-series models is the state space form, with the state of the system representing the various unobserved components, such as trend, cyclical or seasonal fluctuations. Once in state space form (SSF), the Kalman filter provides the means of updating the state, as new observations become available. Once a model is estimated, its suitability can be assessed using goodness fit statistics. Gram area (‘000 Mill. ha) and production (‘000 MT) in India data for the period of 1950-51 to 2011-2012 were analyzed by structural time series model used to forecast for five leading years. II. MATERIALS AND METHODS The study mainly confined to Gram area and production of India. The secondary data of area and production on Gram of 62 year were collected for period 1950-51 to 2011-12. Data collected from Department of Agriculture and Cooperation, Government of India were subjected to analyze through structural time series model. The data are analyzed by using software like MS-EXCEL and Statistical Analysis System (SAS). Structural time series model was adopted to observe the forecast model, the model used was: Structural time Series Model for trend: A structural time series model is set up in term of its various components, like trend, cyclic fluctuations and seasonal variation, i.e. Y t = T t + C t + S t + Ԑ t ... (1) Where Y t is the observed time-series at time t, T t , C t , S t , Ԑ t , are the trend, cyclical, seasonal and irregular components. Abstract : A univariate structural time series model based on the traditional decomposition into trend, seasonal and irregular components is defined. Purpose of present paper is to discuss STM methodology utilized for modelling time-series data in the present of trend, seasonal and cyclic fluctuations. Structural time series model are formulated in such a way that their components are stochastic, i.e. they are regard as being driven by random disturbances. A number of methods of computing maximum Likelihood estimators are then considered. These include direct maximization of various times domain likelihood function. Once a model is estimated, its suitability can be assessed using goodness fit statistics and model used to predict for five leading years. In our study the model developed for gram production, from the forecasting available. Key Words: Structural time series model, forecast, Kalman filter, goodness of fit