The objectives of this research are to propose a Bayesian model for spatial time series analysis to apply the proposed model to forecast a monthly rubber yield in Southern provinces of Thailand and to compare the performance of the proposed model with the classical Holt-Winters Additive Exponential smoothing. The proposed model is a linear mixed model (LMM) with spatial effects which follow a conditional autoregressive model (CAR). Dummy variables are used for seasonal effects. A Bayesian method is used for parameter estimation. The estimated monthly yields are used for the monthly rubber yield forecasting. The dependent variables are the rubber yield in each month of each province. The data are secondary data at a provincial level. The factors considered are spatial effects heterogeneity effects and seasonal effects. The results show that the factors influencing on the amount of rubber yields are spatial heterogeneity and seasonal effects. The proposed model is proper and forecast accurately. Using the mean absolute error (MAE) the proposed model has a better performance compared to the classical Holt-Winters Additive Exponential smoothing in both model fitting and model validating. The proposed model should be the first consideration for spatial time series forecasting.

The objective of this research is to select a proper cumulative function for trend in an analysis of simultaneous multiple time series data with autocorrelation trend and outliers using a Bayesian method. The proposed model is applied to fluctuate stock price data in the stock exchange of Thailand. The data consist of the closing prices of 5 stocks. The algorithms are written in OpenBUGS and the performance of the model is evaluated via a simulation study in R. The cumulative distribution functions considered here are a Weibull cumulative distribution function and an exponential cumulative distribution. They are compared using RMSE MAPE and MAE as criteria. The research found that an exponential cumulative function for trends gives smaller values of the three criteria in both model fitting and model validation for the4 stocks EA NWR and KTB. However a Weibull cumulative distribution function gives smaller values of three criteria for the PTT in both model fitting and model validation.

The objective of this research is to design forecasting support models on demand of durian for domestic markets and export markets for durian gardeners durian entrepreneurs of domestic markets and export markets and the Office of Agricultural Economics to plan durian demand conforming with domestic markets and export markets because of Production and marketing main problems (The Agricultural Information Center The office of Agricultural Economics: 2008) such as over - much durian production during production capital increased farmers durian sold price tended to be lower and lower inefficient domestic market management and more export but cheaper price therefore after that to design forecasting for the purpose in order that durian would not be over demand which is containing factors of the demand of durian fresh durian frozen durian paste and durian chips for domestic markets and export markets exported to Asia America Australia and Africa. The research involves historical collection data in the period of 2002 to 2008. The basic of forecast models are the designed and improved models using an intelligent knowledge-based approach beginning Moving average Deseasonalized Exponential smoothing Double exponential smoothing and Artificial neural network (ANNs) program within the model of new value Creation and comparative accuracy models. The evaluation result of forecasting durian demand showed that lowest MAPE of Deseasonlized models at 3 month was durian fresh durian frozen and durian paste but durian chips showed that lowest MAPE of ANNs at 4 input 10 hidden layers and 1 output was high accurate forecasting and optimal models. The analysis recommendations and error forecast of durian demand are also presented.

The problem of this research is automotive part companies do not have order because of the world economic crisis and effect to automotive industry. Automotive and automotive parts demand in the world decrease. Automotive Industry of Thailand received affect. Automotive part companies cannot export to foreign markets. The companies in Thailand need the forecasting of automotive part quantity demanded by foreign markets by selecting the most accurate model. This paper presents the modeling for forecasting demand for automotive part of foreign markets to Thailand with time series forecasting techniques. In this study the automotive part are wheels and including parts and accessories because Thailand exported the most in group of automotive parts. These five countries have consistently demanded the largest number of wheels and including parts and accessories are Japan China South Korea Germany and Indonesia. The research involves historical collecting data for the period of 1997 to 2008. The evaluation function of forecast error is used with MAPE. The results reveal that each country suits for each model in term of the chosen dimension. The government can use the forecasting demand for effective production planning by informing manufactures.

This research aimed to design and construct a forecasting support system on demand of fresh durian of Thailand by applying 2 mathematical models which were Time Series model and Artificial Neural Network model in order to find an accurate model of forecasting support system on demand of durian yield to forecast durian yield quantity of Thailand in advance and use the forecast results to plan and determine quantity of fresh durian and durian product conforming with domestic consumption and export in order not to be over demand and cheap durian sold price. The forecast models were applied with Time Series model by using Moving Average Weighted Moving Average Single Exponential Smoothing and Holt’s Linear Exponential Smoothing techniques and Back Propagation of Artificial Neural Network model which used 4 input variables relating to durian yield. All of the forecast models were used to forecast total durian yield quantity in 26 provinces of durian cultivated area all around Thailand. Durian yield quantity data used was from 1996 to 2008. After that the forecast results were used to find the errors of MAE MSE RMSE and MAPE. Then the errors of each forecast model were compared. The forecast model which had the least error was the most accurate forecast model. The research result is following. Back Propagation with 4-8-1 structure of Artificial Neural Network model had the least error so that it was the most accurate forecast model to forecast fresh durian yield quantity in advance.