Setting up of the Modem Portfolio Theory in 1952, and later the CAPM model in 1964, initiated a significant scientific research impulse for establishing a model that could be used to forecast the future value of shares in the financial market. A number of models that followed herein were perhaps theoretically more precise. or closer to reality in their assumptions. but also more complicated to calculate. The problem that appeared at the same time. and which has still not been resolved, are certain anomalies which appeared in the capital market that the respective models could not explain. At that point, behavioral finance appears which seeks to explain such anomalies in terms of psychological relationships between the participants in the capital market. This research will try to indirectly test the behavioral elements in the Croatian capital market, following a series of data from 2002 until 2016, with the corresponding sub-periods. It will be done by determining the deviation of the theoretical values that shares should yield, according to the CAPM model, from the real yields that they actually achieved. The hypothesis is that the environment in the capital market will have a significant impact on the applicability of the model for the evaluation of shares. The scientific contribution will be achieved in terms of establishing additional preconditions for the applicability of this model on developing capital markets. This study demonstrated that the SML equation was statistically significant only in 2007, while in all the other years it could not be used to identify wrongly valued shares. Such results confirmed the existence of behavioral elements in the Croatian capital market and its weak relationship with the real economy.