We have done some research to try and locate the significant factors for long/short strategies on US Equities using Morning Star data. We wanted to make a demonstration of how you can use fundamental data in the research environment.

From the research we found that the BookValueYield (BookValuePerShare / Price), PE Ratio and EVToEBITDA are the most significant factors that can explain the return. But BookValueYield is negatively correlated with the stock return but PE Ratio and EVToEBITDA are positively correlated with return. 

The inverse correlation of the BookValueYield factor could be indicative of "affordable" companies; explaining why the negative correlation generates better returns. 

Both the loss probability and the win probability are over 0.5. The excess return for win portfolio is positive and the excess return for loss portfolio is negative.

In future research we could test more factors and test significant ones as independent variables in multi-factor models. Or experiment with long/short strategies, by ranking the stocks according to the significant factors to generate the positive returns.