Equation

pmcore::prelude::simulator::equation

Trait Equation 

pub trait Equation:
    EquationPriv
    + 'static
    + Clone
    + Sync {
    // Required methods
    fn estimate_likelihood(
        &self,
        subject: &Subject,
        support_point: &Vec<f64>,
        error_models: &AssayErrorModels,
    ) -> Result<f64, PharmsolError>;
    fn estimate_log_likelihood(
        &self,
        subject: &Subject,
        support_point: &Vec<f64>,
        error_models: &AssayErrorModels,
    ) -> Result<f64, PharmsolError>;
    fn kind() -> EqnKind;

    // Provided methods
    fn estimate_predictions(
        &self,
        subject: &Subject,
        support_point: &Vec<f64>,
    ) -> Result<Self::P, PharmsolError> { ... }
    fn nouteqs(&self) -> usize { ... }
    fn nstates(&self) -> usize { ... }
    fn simulate_subject(
        &self,
        subject: &Subject,
        support_point: &Vec<f64>,
        error_models: Option<&AssayErrorModels>,
    ) -> Result<(Self::P, Option<f64>), PharmsolError> { ... }
}
Expand description

Trait for model equations that can be simulated.

This trait defines the interface for different types of model equations (ODE, SDE, analytical) that can be simulated to generate predictions and estimate parameters.

§Likelihood Calculation

Use estimate_log_likelihood for numerically stable likelihood computation. The deprecated estimate_likelihood is provided for backward compatibility.

Required Methods§

fn estimate_likelihood( &self, subject: &Subject, support_point: &Vec<f64>, error_models: &AssayErrorModels, ) -> Result<f64, PharmsolError>

👎Deprecated since 0.23.0: Use estimate_log_likelihood() instead for better numerical stability

Estimate the likelihood of the subject given the support point and error model.

Deprecated: Use estimate_log_likelihood instead for better numerical stability, especially with many observations or extreme parameter values.

This function calculates how likely the observed data is given the model parameters and error model. It may use caching for performance.

§Parameters
  • subject: The subject data
  • support_point: The parameter values
  • error_model: The error model
§Returns

The likelihood value (product of individual observation likelihoods)

fn estimate_log_likelihood( &self, subject: &Subject, support_point: &Vec<f64>, error_models: &AssayErrorModels, ) -> Result<f64, PharmsolError>

Estimate the log-likelihood of the subject given the support point and error model.

This function calculates the log of how likely the observed data is given the model parameters and error model. It is numerically more stable than estimate_likelihood for extreme values or many observations.

Uses observation-based sigma, appropriate for non-parametric algorithms. For parametric algorithms (SAEM, FOCE), use crate::ResidualErrorModels directly.

§Parameters
  • subject: The subject data
  • support_point: The parameter values
  • error_models: The error model
§Returns

The log-likelihood value (sum of individual observation log-likelihoods)

fn kind() -> EqnKind

Provided Methods§

fn estimate_predictions( &self, subject: &Subject, support_point: &Vec<f64>, ) -> Result<Self::P, PharmsolError>

Generate predictions for a subject with given parameters.

§Parameters
  • subject: The subject data
  • support_point: The parameter values
§Returns

Predicted concentrations

fn nouteqs(&self) -> usize

Get the number of output equations in the model.

fn nstates(&self) -> usize

Get the number of state variables in the model.

fn simulate_subject( &self, subject: &Subject, support_point: &Vec<f64>, error_models: Option<&AssayErrorModels>, ) -> Result<(Self::P, Option<f64>), PharmsolError>

Simulate a subject with given parameters and optionally calculate likelihood.

§Parameters
  • subject: The subject data
  • support_point: The parameter values
  • error_model: The error model (optional)
§Returns

A tuple containing predictions and optional likelihood

Dyn Compatibility§

This trait is not dyn compatible.

In older versions of Rust, dyn compatibility was called "object safety", so this trait is not object safe.

Implementors§