In this study, an analysis of the monthly mean values of the fraction of the sky covered by clouds (C) of all types and the duration of bright sunshine hours (S) and global solar radiation on the horizontal surface (H) was carried out to establish relationship for solar radiation estimation for Bangladesh. The relationship between C and (1 − S/S 0 ), where S 0 is the maximum possible hours of sunshine, was found to be nonlinear. The relationship between C and (1 − H/H 0 ), where H 0 is the extraterrestrial solar radiation, was also found to be nonlinear. Few quadratic regression equations were derived for different regions of Bangladesh. The data were separated in Zone 1 (below the tropic of cancer, i.e., ≤23.5°N) and Zone 2 (above the tropic of cancer, i.e., ≥23.5°N). All the developed models were checked statistically. It was found that it is always better to estimate solar radiation from sunshine duration as it gives the best estimation and accuracy, but if there is no sunshine duration data available for a site, then solar radiation can be estimated directly from cloud cover data. Estimating sunshine duration first from cloud data to estimate solar radiation introduces more error as it has to go through the modeling stage twice. In case of the need of sunshine duration data only, the estimation from cloud cover is quite helpful. It was also found from the study that for small country like Bangladesh separating data according to latitude has insignificant effect.