Media news reveals soft information about economic linkages between firms that is not immediately incorporated into stock price. In this paper, we propose a novel measure for aggregate market risk based on news network and news tones, namely media connection index (MCI). We show that the change of MCI predicts a negative return with a higher in-sample and out-of-sample performance than average correlation index in Pollet and Wilson (2010). In addition, our findings are also statistically as well as economically significant even though we control for different economic predictors used in Goyal and Welch (2008). We also find this measure is capable of predicting cross-sectional stock returns sorted by B/M, size, momentum, industry, investment and profitability. Further analysis shows that the predictability of our measure mainly comes from the negative tones which is consistent with Tetlock (2007).
↧