In part I, we presented a number of issues that are relevant to the context of prediction model development and application. We summarize these issues as preliminaries for model development in a proposed checklist. In part II, we focus on the development of prediction models that are valid for the population from which the sample originates. Generalizability to other, plausibly related, populations is discussed in Part III. We will discuss seven steps of model development in the following chapters (Chaps. 7-18).
Steyerberg, E. W. (2009). Dealing with missing values (pp. 115–137). https://doi.org/10.1007/978-0-387-77244-8_7