Stock prices are typically modeled using a lognormal distribution. This is because stock prices cannot fall below zero (they are bounded by zero), which aligns with the properties of the lognormal distribution. The underlying assumption is that the continuously compounded returns (log returns) of the stock are normally distributed.