Abstract-Predicting the efficiency of solar photovoltaic and solar thermal systems needs global solar radiation data and its estimation models. Furthermore, daily global solar radiation models are required for engineering design and planning projects. In this paper, five daily global solar radiation models for tilted surfaces are compared and analyzed. They are namely: Jordan, Hay, Baltas, Perez and Klucher models. In order to assess the applicability and accuracy of the models proposed in this study, the radiometric and geographical data are used. Comparative analyses are established according to 50 years meteorological data. They are supplied by the National Office of Meteorology -NOM-, (Dar El Beida, Algeria, 3.21° E, 36.68°N) and the Unit for Applied Research in Renewable Energy -UARRE-(Ghardaïa, Algeria, 3.80 °E, 32.4°N). This latest selected station is quite interesting as it is situated a hundred kilometers far from the famous hybrid (solar, gas) station of Hassi R'Mel). Moreover, it provides reliable data. Finally, we find it necessary to mention that the obtained results show that the selected models have been able to estimate global solar radiation accurately so that complex equation types could not obviously improve the precision. The adequacy of a model is established on the basis of various usual errors such as the root mean square error (RMSE), the mean bias error (MBE) and the mean percentage error (MPE). Results are with attractive margins of error and in accordance with experimental measurements. These models take into account the type of contributions to diffuse radiation, that is to say, the isotropic and anisotropic sky.