For the climatic variables solar irradiance and ambient temperature, a method has been devised for identifying the important cyclical components. After the contribution of the steady periodic part has been removed, the residual time series are analysed. It is shown that the daily solar irradiance residuals are a first-order autoregressive process while the daily average ambient temperature residuals are a third-order autoregressive process. This is independent of location. These results plus a knowledge of the characteristics of the cyclical components enable one to devise a methodology for generating synthetic series for these variables which are statistically indistinguishable from the original time series. Using this methodology, one can generate input data for models of heat hows in houses or solar hot water systems or other devices.