Date of Publication :14th March 2018

Abstract: The present study was conducted with the prime objective to develop a stochastic time series model for prediction of ground water fluctuation in Kaushambi. The area of the Kaushabi is 2012.8 sq kms and the data of 10 years from 2002 to 2011 were used to developed the model Autoregressive (AR) models of orders 0, 1 and 2 were tried to develop a autoregressive time series model for annual ground water fluctuation. The various parameters of model were estimated by the general recursive formula proposed by Kottegoda (1980).The goodness of fit and adequacy of models were tested by Box-Pierce Portmanteau test, Akaike Information Criterion and by comparison of measured and predicted correlogram. The AIC value for AR (1) model is lying between AR (0) and AR (2) which is satisfying the selection criteria. The mean prediction error is less in case of ground water fluctuation by AR (1) model. The Graphical representation between measured and generated correlogram is show that there is close agreement between pre and post monsoon. The comparison between the measured and predicted ground water table and pre and post monsoon by AR (1) model, clearly shows that the developed model can be use efficiently for the prediction of pre and post at the kaushambi district Ground water table

Reference :