Do analysts overreact to extreme good news in earnings?

We provide an alternative explanation for the previous finding of analysts' overreaction to extreme good news in earnings. We show that such finding could be a result of analysts' rational behavior in the face of high earnings uncertainty rather than their cognitive bias. Extreme earnings performance tends to be associated with higher earnings uncertainty that generally leads to more forecast optimism. Once this effect is accounted for, the univariate result of analysts' overreaction to extreme good news in earnings is subsumed, leaving only their underreaction in general. © 2007 Springer Science+Business Media, LLC