Suppose you hold a stock and are considering whether to sell it or keep it. You initially believe that the probability the stock will rise in value in the long run is 0.7. You decide you will sell the stock when the probability of a long run drop in value reaches 0.5. Then, over the course of time you watch the changes in the value of the stock day today, each day's outcome serving as a forecast of future value. Further, suppose that P(Rise | rise) = 0.6 where RISE indicates a long-run future rise in value and rise indicates an observed daily rise in value. Correspondingly, P(FALL | fall) = 0.6, , where FALL indicates a long-run decline in future values of the stock and fall indicates a daily observed decline in value. Suppose the probability that you misperceive a signal given it contradicts your current belief is q = 0.4. How many daily declines would you need to observe before you would sell the stock according to Rabin and Schrag's model of confirmation bias? What would a Bayesian's beliefs be regarding the probability of a decline at that point? How many daily declines would you need to observe in order to sell if you had perfect perception (q = 0)?