Tutorials › Pharmacometrics › Additive, Proportional, and Combined Error Models
Pharmacokinetics · Population PK

Additive, Proportional, and Combined Error Models

Learn how residual error models describe the difference between observed and model-predicted concentrations—and when additive, proportional, and combined error structures are useful in pharmacokinetic modeling.

Intermediate Population PK Residual Variability PK Modeling
01 · The big picture

1. What Is a Residual Error Model?

In pharmacokinetic modeling, the structural model describes the systematic behavior of drug concentrations over time. Even when the structural model is appropriate, observed concentrations will not lie exactly on the model-predicted curve.

The difference between an observation and its corresponding model prediction is represented through a residual error model, also called an observation model.

Time C Model prediction Observed concentrations

A residual error model describes the remaining discrepancy between observed concentrations and the concentrations predicted by the structural PK model.

Core idea: the structural model describes the expected concentration, while the residual error model describes how observations vary around that prediction.
02 · Observation model

2. The Observation Model

Let \(C_i\) denote the observed concentration for observation \(i\), and let \(F_i\) denote the concentration predicted by the structural PK model.

A general observation model can be written conceptually as:

\[ C_i = \text{systematic prediction} + \text{residual error} \]

For an additive model, this becomes:

\[ C_i = F_i + \epsilon_i \]

The residual term \(\epsilon_i\) represents unexplained variability between the observed concentration and the model prediction.

The important point is that residual variability is not the same thing as between-subject variability. Between-subject variability describes differences in underlying PK parameters among individuals. Residual variability describes observation-level discrepancies that remain after accounting for the structural model and modeled subject-level variability.

ComponentWhat it describesTypical example
Structural model Systematic concentration-time behavior Clearance, volume, absorption rate
Interindividual variability Differences between individuals in PK parameters One patient has higher CL than another
Residual error Observation-level unexplained variability Assay variation or model discrepancy
03 · Additive error

3. The Additive Error Model

The additive error model assumes that the magnitude of the residual error is approximately constant across the concentration range.

\[ C_i = F_i + \epsilon_i \] $$ \epsilon_i \sim N(0,\sigma^2) $$

Here, \(\sigma\) is the residual standard deviation. The model therefore assumes that the standard deviation of the observation error does not depend on the magnitude of the predicted concentration.

What does "additive" mean?

Suppose the residual standard deviation is \(1\) mg/L. Under an additive model, a prediction of \(2\) mg/L and a prediction of \(20\) mg/L would both have approximately the same residual standard deviation of \(1\) mg/L.

Predicted concentrationApproximate residual SDInterpretation
2 mg/L1 mg/LError is large relative to the prediction
10 mg/L1 mg/LError is moderate relative to the prediction
20 mg/L1 mg/LError is relatively small

Thus, an additive error model implies a constant absolute error scale, not a constant relative error scale.

When it can be useful: additive error can be reasonable when assay or measurement variability is approximately constant in absolute concentration units, particularly when concentrations remain within a relatively narrow range.
04 · Proportional error

4. The Proportional Error Model

The proportional error model assumes that residual variability increases with the magnitude of the predicted concentration.

\[ C_i = F_i(1+\epsilon_i) \] $$ \epsilon_i \sim N(0,\sigma^2) $$

Equivalently, the residual component can be expressed as a standard deviation proportional to the prediction:

\[ SD(C_i\mid F_i)=\sigma F_i \]

The key feature is that the coefficient of variation is approximately constant.

What does "proportional" mean?

Suppose \(\sigma=0.20\). A prediction of \(5\) mg/L would have an approximate residual SD of \(1\) mg/L, while a prediction of \(20\) mg/L would have an approximate residual SD of \(4\) mg/L.

\[ SD(C_i\mid F_i)=0.20F_i \]
Predicted concentrationApproximate residual SDRelative error scale
5 mg/L1 mg/L20%
10 mg/L2 mg/L20%
20 mg/L4 mg/L20%

This makes proportional error particularly useful when measurement variability is naturally expressed as a percentage of concentration rather than as a fixed concentration amount.

Key distinction: additive error assumes approximately constant absolute variability, whereas proportional error assumes approximately constant relative variability.
05 · Combined error

5. The Combined Additive + Proportional Error Model

Many PK datasets show a pattern in which variability is not well described by either a purely additive or purely proportional model across the entire concentration range.

A combined error model allows both components to contribute:

\[ C_i = F_i + F_i\epsilon_{1i}+\epsilon_{2i} \]

where the proportional component is represented by \(F_i\epsilon_{1i}\), while \(\epsilon_{2i}\) represents an additive component.

A common formulation is:

\[ \epsilon_{1i}\sim N(0,\sigma_{\mathrm{prop}}^2), \qquad \epsilon_{2i}\sim N(0,\sigma_{\mathrm{add}}^2) \]

Under the usual independence assumption, the conditional variance is approximately:

\[ Var(C_i\mid F_i) = \sigma_{\mathrm{add}}^2 + \sigma_{\mathrm{prop}}^2F_i^2 \]

Therefore, the residual standard deviation changes with the predicted concentration rather than remaining completely constant or increasing strictly in direct proportion.

Additive Proportional Combined Predicted concentration Residual SD

Conceptual residual standard deviation as a function of predicted concentration. Additive error is approximately constant; proportional error increases with prediction; combined error incorporates both patterns.

Why combine them? A combined model can accommodate a baseline amount of absolute error at low concentrations while allowing variability to increase with concentration at higher values.
06 · Comparing models

6. Additive vs. Proportional vs. Combined Error

FeatureAdditiveProportionalCombined
Basic form \(C=F+\epsilon\) \(C=F(1+\epsilon)\) \(C=F+F\epsilon_1+\epsilon_2\)
Absolute error Approximately constant Increases with \(F\) Can have both components
Relative error Increases as concentration decreases Approximately constant Can change across concentration range
Low concentrations Can be useful May become problematic Can accommodate additive noise
High concentrations May underestimate variability Can be useful Can accommodate proportional variability
Main assumption Constant absolute SD Constant relative SD Both absolute and relative components

These are not simply three levels of model complexity. They represent different assumptions about how observation variability behaves across the concentration range.

07 · Variance behavior

7. Understanding the Variance of Each Error Model

One of the clearest ways to understand residual error models is to examine how their variance changes with the model prediction.

Additive model

\[ Var(C_i\mid F_i)=\sigma_{\mathrm{add}}^2 \]

The variance is constant regardless of \(F_i\).

Proportional model

\[ Var(C_i\mid F_i)=\sigma_{\mathrm{prop}}^2F_i^2 \]

The variance increases quadratically with the predicted concentration.

Combined model

\[ Var(C_i\mid F_i) = \sigma_{\mathrm{add}}^2 + \sigma_{\mathrm{prop}}^2F_i^2 \]

The combined model therefore has a constant variance component plus a concentration-dependent component.

Useful interpretation: the combined model can be viewed as transitioning from predominantly additive behavior at low concentrations toward increasingly proportional behavior as concentrations become larger.
08 · Relative error

8. Why the Same Error Can Look Different at Different Concentrations

Consider an absolute residual of \(1\) mg/L.

Observed or predicted concentrationAbsolute errorRelative error
2 mg/L1 mg/L50%
5 mg/L1 mg/L20%
10 mg/L1 mg/L10%
20 mg/L1 mg/L5%

The same absolute measurement error therefore represents a very different percentage of the concentration depending on where the observation lies in the concentration range.

This is one reason residual error models should be evaluated against the concentration range and measurement process rather than selected solely because one model has a simpler equation.

09 · Log-transformed models

9. How Log Transformation Changes the Error Model

Another common approach is to model concentrations on the logarithmic scale.

Suppose:

\[ \log(C_i)=\log(F_i)+\epsilon_i \]

Exponentiating both sides gives:

\[ C_i=F_i e^{\epsilon_i} \]

This creates a multiplicative error structure on the original concentration scale.

Log-scale modeling can therefore be useful when variability is naturally related to relative rather than absolute concentration differences. It also ensures that the back-transformed prediction remains positive.

Important: an additive error model on the log scale is not the same as an additive error model on the original concentration scale. The scale on which the residual model is defined matters.
10 · Model selection

10. How Should You Choose an Error Model?

Error-model selection should be guided by the observed data, the assay or measurement process, the concentration range, and model diagnostics.

  1. Start with scientific knowledge. Consider whether measurement error is expected to be approximately constant in absolute units or proportional to concentration.
  2. Inspect the concentration range. A narrow range may make different residual models difficult to distinguish.
  3. Fit plausible candidate models. Compare additive, proportional, combined, or log-scale alternatives when scientifically reasonable.
  4. Examine residual diagnostics. Look for trends in residuals versus predictions, time, or other relevant variables.
  5. Evaluate parameter estimates. Check whether the residual parameters are plausible and reasonably estimated.
  6. Consider objective-function or likelihood-based comparisons where appropriate. Statistical comparison can support, but should not replace, examination of model adequacy.
  7. Prefer an adequate and interpretable model. Additional complexity should have a modeling purpose.
Model-selection principle: the goal is not automatically to choose the most flexible residual error model. The goal is to adequately characterize the observation variability without introducing unnecessary complexity.
11 · Diagnostics

11. What Do Residual Diagnostics Tell Us?

Residual diagnostics are particularly important because a residual error model can appear mathematically reasonable while still leaving systematic patterns in the data.

Residuals versus predictions

A plot of residuals against predictions can reveal whether residual variability changes systematically with concentration.

For example, a funnel-shaped pattern may indicate that variability increases as predicted concentrations increase, suggesting that a purely additive error model may not adequately describe the data.

Conditional weighted residuals

Population PK analyses often use residual diagnostics such as conditional weighted residuals to examine whether residuals are approximately centered around zero and whether systematic trends remain.

Residuals versus time

Patterns over time can indicate that the structural PK model is not adequately capturing some feature of the concentration-time profile. Not every residual pattern should be "fixed" by changing the error model.

Key distinction: systematic residual trends can arise from an inadequate structural model, an inadequate covariate model, an inappropriate residual error model, or combinations of these. Residual diagnostics should therefore be interpreted in context.
12 · Worked example

12. Worked Example: Comparing Additive and Proportional Error

Suppose a population PK model predicts a concentration of 10 mg/L.

Step 1: Additive error

Assume the additive residual standard deviation is \(2\) mg/L.

\[ SD(C\mid F)=\sigma_{\mathrm{add}}=2\text{ mg/L} \]

The model therefore implies a residual SD of \(2\) mg/L at a prediction of \(10\) mg/L.

Step 2: Proportional error

Now suppose the proportional residual standard deviation is represented by a coefficient of variation of \(20\%\), or \(\sigma_{\mathrm{prop}}=0.20\).

\[ SD(C\mid F)=0.20(10)=2\text{ mg/L} \]

At \(10\) mg/L, the additive and proportional models happen to produce the same residual SD.

Step 3: Compare them at 2 mg/L

Under the additive model:

\[ SD(C\mid F)=2\text{ mg/L} \]

Under the proportional model:

\[ SD(C\mid F)=0.20(2)=0.4\text{ mg/L} \]

The models now make very different assumptions. The additive model retains a \(2\) mg/L residual SD, whereas the proportional model predicts a much smaller \(0.4\) mg/L residual SD.

Step 4: Compare them at 20 mg/L

Under the additive model:

\[ SD(C\mid F)=2\text{ mg/L} \]

Under the proportional model:

\[ SD(C\mid F)=0.20(20)=4\text{ mg/L} \]

The proportional model now allows twice as much absolute variability as the additive model.

Step 5: Interpretation

PredictionAdditive SDProportional SD
2 mg/L2 mg/L0.4 mg/L
10 mg/L2 mg/L2 mg/L
20 mg/L2 mg/L4 mg/L

This example shows why the two models can behave similarly over one part of the concentration range while making very different assumptions at the extremes.

13 · Combined model

13. Worked Example: Understanding a Combined Error Model

Suppose a combined error model has:

\[ \sigma_{\mathrm{add}}=0.5\text{ mg/L} \] $$ \sigma_{\mathrm{prop}}=0.20 $$

The conditional variance is:

\[ Var(C\mid F) = 0.5^2+(0.20F)^2 \]

At \(F=2\) mg/L

\[ SD(C\mid F) = \sqrt{0.5^2+(0.20\times2)^2} \] $$ = \sqrt{0.25+0.16} \approx0.64\text{ mg/L} $$

At \(F=10\) mg/L

\[ SD(C\mid F) = \sqrt{0.5^2+(0.20\times10)^2} \] $$ = \sqrt{0.25+4} \approx2.06\text{ mg/L} $$

At \(F=20\) mg/L

\[ SD(C\mid F) = \sqrt{0.5^2+(0.20\times20)^2} \] $$ = \sqrt{0.25+16} \approx4.03\text{ mg/L} $$

The additive component provides a baseline amount of variability, while the proportional component becomes increasingly important as the predicted concentration increases.

14 · Common mistakes

14. Common Error-Modeling Mistakes

Mistake 1: Confusing residual error with IIV

Residual error describes observation-level variability. Interindividual variability describes differences in PK parameters among individuals. They represent different levels of the model.

Mistake 2: Assuming proportional error means constant absolute error

Under proportional error, the absolute residual SD increases with the predicted concentration. What remains approximately constant is the relative error scale.

Mistake 3: Automatically choosing combined error

A combined model is flexible, but flexibility alone is not evidence that it is appropriate. If the data support a simpler error structure, unnecessary complexity may not provide a meaningful benefit.

Mistake 4: Treating residual patterns as purely an error-model problem

A trend in residuals can result from misspecified clearance, volume, absorption, covariates, time structure, or other aspects of the structural model. Changing the residual error model may not solve the underlying problem.

Mistake 5: Ignoring the concentration scale

Error behavior should be interpreted on the scale on which the observation model is defined. An error model on the original concentration scale has different implications from one defined after logarithmic transformation.

Mistake 6: Assuming one error model works for every assay

Residual variability depends partly on how observations are generated and measured. The appropriate model can therefore differ among compounds, assays, laboratories, concentration ranges, and study designs.

15 · Population PK implementation

15. How Error Models Appear in Population PK

In population PK software, residual error models are typically specified alongside the structural model and interindividual variability model.

A simplified hierarchy can be represented as:

\[ \text{Dose} \rightarrow \text{Structural PK Model} \rightarrow F_i \rightarrow \text{Residual Error Model} \rightarrow C_i \]

For an individual subject \(i\), the structural model may depend on individual-specific parameters such as:

\[ CL_i,\quad V_i,\quad k_{a,i} \]

Those parameters can incorporate interindividual variability. The residual error model is then applied to the resulting prediction.

Model layerQuestion
Structural modelWhat concentration-time behavior is expected?
IIV modelHow do PK parameters differ among individuals?
Covariate modelCan observed characteristics explain some parameter differences?
Residual error modelHow do observations vary around the individual predictions?

This separation is fundamental to interpreting population PK models correctly.

16 · Practical interpretation

16. Reading Residual Patterns in Practice

Consider several conceptual patterns that might appear in a residual-versus-prediction plot.

PatternPossible interpretation
Approximately constant spread An additive error model may be plausible.
Spread increases with prediction A proportional component may be needed.
Large spread at low concentrations plus increasing spread at high concentrations A combined model may be plausible.
Systematic positive or negative trend Potential structural-model or covariate misspecification should be investigated.
Strong asymmetry or non-normal residual behavior The assumed residual distribution or observation scale may require reconsideration.

These patterns are diagnostic clues rather than automatic decisions. A residual plot should be interpreted alongside the study design, assay characteristics, structural model, and other diagnostics.

17 · Practical workflow

17. A Practical Error-Modeling Workflow

  1. Understand the measurement process. Determine whether assay variability is expected to be approximately absolute, proportional, or mixed.
  2. Inspect the concentration range. Very wide ranges can make the distinction between additive and proportional variability especially important.
  3. Fit a scientifically plausible starting model. Do not begin with maximum complexity automatically.
  4. Inspect residual diagnostics. Examine residual behavior versus predictions and time.
  5. Compare alternative error structures. Consider additive, proportional, combined, and log-scale formulations where appropriate.
  6. Check parameter precision and plausibility. A complicated residual model should still be identifiable and interpretable.
  7. Evaluate the overall model. Residual error should be considered together with structural model adequacy, IIV, and covariate relationships.
  8. Select the simplest adequate formulation. The final model should represent the observation process sufficiently well for the scientific purpose.
Practical rule: use residual diagnostics to understand what the data are saying, but do not use the residual error model as a substitute for getting the structural PK model right.

18. Key Takeaways

  • The residual error model describes unexplained variability between observed concentrations and model predictions.
  • An additive error model assumes approximately constant absolute residual variability.
  • A proportional error model assumes that absolute variability increases with predicted concentration while relative variability remains approximately constant.
  • A combined additive + proportional model allows both a baseline absolute component and a concentration-dependent component.
  • The variance of a combined model can be represented as the sum of additive and proportional variance components under the usual independence assumption.
  • Residual error is distinct from interindividual variability: residual error describes observation-level discrepancies, whereas IIV describes differences in PK parameters among individuals.
  • Log-scale models create multiplicative error behavior on the original concentration scale.
  • Residual-versus-prediction diagnostics can help identify whether variability changes across the concentration range.
  • Systematic residual patterns are not necessarily solved by changing the error model; structural-model and covariate misspecification should also be considered.
  • The most appropriate error model depends on the measurement process, concentration range, data, diagnostics, and scientific purpose.
  • A more complex combined error model is not automatically preferable to a simpler model.
Next step

Where to Go Next

A natural progression is to study interindividual variability in population PK, followed by interoccasion variability, residual unexplained variability models, and the relationship between fixed effects, random effects, and residual error.

These concepts together provide the foundation for understanding how a population PK model separates typical population behavior, between-subject differences, and observation-level variability.

← Back to Pharmacokinetics Tutorials