Observations

Observations are the link between the model and the data. The model describes the relationships between covariates, parameters, and variables. The data represent a random sampling of the system that the model describes. The various types of observation statements that are available in the Pharmacometrics Modeling Language serve to build a statistical structure around the likelihood of a given set of data.

Observation statements are used to build the likelihood function that is maximized during the modeling process. The observation statements indicate how to use the data in the context of the model.

Observe statement for Gaussian Residual models

LL statement for user-defined log-likelihood models

Count statement for Count models

Event statement for Time-to-event models

Multi statement for Categorical models

Ordinal statement for Ordinal Responses

Observation statement action codes

The model can contain any number of each of the observation statements above and any combination of these statements as needed.