MDL used in defining the model is intended more as a descriptive language rather than a programmatic one. The user is free to define any model, however they must also be aware that the specified model may not be useable with all target software.Īs stated in the Introduction, the Model Object is intended to convey the mathematical and statistical definitions required to completely define the model. It should be noted however that MDL does not guide the user about whether the model that is defined is suitable for a given purpose or for any target software. The Model Object should also be independent of the data – where possible we use enumerated types for categorical covariates and outcomes so that definition of the model is clear to any user regardless of the data used in a given task. The same model should be able to be used for a variety of tasks – estimation, simulation or optimal design without recoding.
The Model Object is intended to specify the model independent of the target software which will be used for the task (estimation or simulation).
MDL defines language elements that allow the user to code a wide variety of models and in a variety of ways. The Model Object within the MDL is intended to describe the mathematical and statistical properties of the model.
9.4.1 How to read the MDL Reference Guide.9.1.5 Attributes, Arguments, Properties and Values.8.3 Mapping of variable names between MDL Objects.6.4.2 Empirical distribution specification with inline data.6.4.1 Non-parametric distribution specification with inline data.6.4 Non-parametric and empirical distributions as priors – inline data.6.3.1 Parametric distributions as priors.6.1 Prior distributions vs initial values vs fixed values.5.3 How are Task Properties used by MDL and PharmML?.5.2 Why use Task Properties for settings and options rather than arguments of functions in the ddmore R package?.4.11 Combining COMPARTMENT and DEQ blocks.4.9.7 INDIVIDUAL_VARIABLES definitions in practice.4.9.6 Conditional assignment of INDIVIDUAL_VARIABLES.4.9.5 INDIVIDUAL_VARIABLES where the variable is defined in the.4.9.4 INDIVIDUAL_VARIABLES without inter-individual variability.4.9.3 Mixed effect model defined by equations.4.9.2 General mixed effect model with Gaussian random effects.4.9.1 Mixed effect model with linear fixed effects and normally distributed.2.2.8 Assignment to multiple variables using define.2.2.7 Assignment to a single variable using variable.2.2.6 Defining model inputs or time-varying covariates.2.2.4 Mapping data variables to model variability levels.2.2.2 Defining the independent variable.1.6 The MDL Integrated Development Environment.
1.5 Task Execution with the ddmore R package.