Research work · 2014
Interpreting the NONMEM Output
Joel S. Owen, Jill Fiedler‐Kelly
Abstract
This chapter discusses the classical estimation methods in NONMEM. It focuses primarily on reviewing the contents and specific types of output of the NONMEM report file. The NONMEM report file is a text file consisting of output from the various components of the NONMEM system relevant to a particular model fitting, in addition to error messages and other useful diagnostic data. There are several types of errors that may occur in NONMEM runs, some of which are described in detail in the chapter. They are syntax (NM-TRAN) errors, estimation step failures, covariance step failures, PREDPP errors, other types of NONMEM errors, and FORTRAN compiler or run-time errors. Finally the chapter provides general Suggestions for diagnosing problems.