Resumen
The present study analyses the effect of trade volume on market analysts? purchase and sell recommendation choices. The research analyses 7,293 consensus recommendations regarding Brazilian listed companies for the period 2008-2014. Sample data includes firms? fundamentalist characteristics, as total assets, return, net income and dividends, with the objective to identify the factors taken under account by analysts for their recommendation evaluations. Applying unbalanced panel data regression strategy, we find that analysts prefer to recommend shares with higher observed trading volume, and the shares with more favourable evaluations are those with higher observed trading volume, which is agreeing with theoretical expectations. Other significative covariates for recommendations are the earnings before interests and taxes ? EBIT, return per share, return of assets ? ROA, paid dividends, and the price/equity ratio.