Two stochastic models for representing the sequence of daily precipitation amounts at a point, the Markov chain model proposed by Haan et al. [1976] and the chain‐dependent process proposed by Katz [1977a, 1977b], are contrasted. Through a further analysis of daily precipitation data, it is shown that there is a lack of evidence to support the superiority of the Markov chain model over the simpler chain‐dependent process.