The objectives of this research are to propose a model for estimating and detecting rice yields in Thailand to investigate the factors related to the rice yields to find the trend and to construct the maps of rice yields in Thailand. A linear mixed model (LMM) including the spatial effects which follow the conditional autoregressive model (CAR) and the time effects which follow the first-order linear autoregressive (AR(1)) model is proposed. The estimated rice yields are used to construct the rice yield maps in Thailand. The dependent variables are the rice yield in each month of each province. The factors considered are rainfall average temperatures and regions. The results indicate that the factors enfluencing the rice yields are rainfall average temperatures and regions (North Northeast South East West Central region is a refrence region). The trend also effects the rice yields. The rice yeilds maps are easy for readers to identify which areas have high or low yields and easy to compare the yields among areas by looking at their different colors.

The objectives of this research are to propose a linear mixed model (LMM) with spatial correlation for an analysis of rice yields in Thailand to estimate the rice yield in each month of all provinces in Thailand to investigate factors influencing on the rice yields and to construct the maps of rice yields. A linear mixed model (LMM) in which the spatial effects follow the conditional autoregressive model (CAR) is used. The estimated rice yields are used to construct the rice yield maps. The dependent variables are the rice yield in each month of each province. The data are secondary data at a provincial level. The factors considered are rainfall averaged temperatures and regions. The results show that the factors influencing on the rice yields are rainfall averaged temperature and region. The amount of regional effects ordering from largest to smallest values are northestern region southern region western region central region eastern region and northern region respectively. The top ten provinces and months with high yields ranking from largest to smallest values are Ubon Ratchathani in November (1328000 ton) Surin in Novermber (1070000 ton) Si Sa Ket in Novermber (1023000 ton) Buri Ram in November (939400 ton) Nakhon Ratchasima in November (907500 ton) Roi Et in Novermber (852800 ton) Chiang Rai in Novermber (769800 ton) Khon Kaen in Novermber (692800 ton) Maha Sarakham in Novermber (676200 ton) Udon Thani in Novermber (620600 ton) respectively. The rice yeilds maps are easy for readers to identify which areas have high or low yields. They are a useful tool for planning decision making and implementing plans to help rice famers to increase their rice production.

The objectives of this research are to propose an approate and efficient model for analyzing rice corn and sugar yields data. A linear mixed model (LMM) with spatial correlation following the conditional autoregressive model (CAR) was adopted. The results show that for rice yields if the amount of rainfall in a harvest season increases 1 mm the yields will decease by 25.74 tons. if the temperature in the harvest season increases 1 Celsius the yields will decease by 212.60 tons and according to the trend the yield will increase by 15.80 tons each year. For sugar cane yields if the amount of rainfall in the harvest season increases 1 mm the yields will decease by 51.35 tons. if the temperature in the harvest season increase 1 Celsius the yields will incease by 121.30 tons and according to the trend the yield will increase by 29.93.80 tons each year. For corn yield if the amount of rainfall in the havest season increases 1 mm the yields will incease by 73.45 tons. if the temperature in he havest season increases 1 celsius the yields will decease by 40.44 tons and according to the trend the yield will increase by 1.74 tons each year.