Statistical models for predicting the solar radiation have been developed. In any prediction of the solar radiation, an
understanding of its characteristics is of fundamental importance. This study presents an investigation of a
relationship between solar radiation and temperature in Perlis, Northern Malaysia for the year of 2006. To achieve
this, the data are presented in daily averaged maximum and minimum air temperature, and daily averaged solar
radiation. Since the scatter plots represent the straight line, the linear regression model was selected to estimate the
solar radiation. It was found that the linear correlation coefficient value is 0.7473 shows that a strong linear
relationship between solar radiation and temperature. The analysis of variance R2 is 0.5585 that is; about 56 percent
of the variability in temperature is accounted for by the straight-line fit to solar radiation. Based on the results, the
fitted model is adequate to represent the estimation of solar radiation.