We are interested in predicting the return on a particular financial asset. The return on the asset at time t is simple the change in its price at time t divided by the price of the previous time period. This is O 1. not a statistical learning problem. 2. both a classifcation and a regression problem because it is not clear whether the return on an asset is a continuous or a discrete random variable. 3. a regression problem because the return on an asset is a continuous random variable. O 4. a classification problem because the sign of an asset's return is a discrete random variable. We are interested to know whether the price goes up or down, and thus the output is a binary random variable (taking the values "up" and "down")