<< JRScience Home Volume 13 No.1 June 2002 << Back to Volume Home
FORECASTS OF A SET OF FUTURE OBSERVATIONS OF PAKISTAN'S PRIVATE CONSUMPTIONS AND INVESTMENTS: A BAYESIAN APPROACH UNDER M-STAGE MODEL
Hayat M. Awan
Institute of Management Sciences, Bahauddin Zakariya University, Multan, Pakistan.
In management and administration, the need for planning and control is important because lead time for managerial decision making ranges from several years or more to a few days or even hours. Information about future events from forecast is usually a critical input into a wide range of managerial and administrative decision-making process. In this paper the problem of forecasting the private consumptions and investments of Pakistan where the prior knowledge may be derived in M-stage in the manner first introduced by Lindley and Smith  have been discussed. The predictive distribution of the future set of observations is derived under the M-stage model and forecast of the private consumptions and investments of Pakistan for 1981-82 to 1987-88 have been computed.
Keywords: Bayesian Inference, Forecasting, M-Stage Model, Informative Prior, Posterior and Predictive Distributions, Predictive Interval, Variance-Covariance Matrix.