Results

Eta Worksheet

Empirical Bayesian estimates of random differences from population expected values for each subject and occasion. This worksheet presents eta estimates by scenario and by model being compared.

Omega Worksheet

Estimates of the random effects variance-covariance matrix, correlation matrix and shrinkage. This worksheet is presented by scenario and by model being compared.

Overall Worksheet

Stacks each model 'Overall' worksheet result but does not present any comparison calculation. Columns include:

Name: Model names being compared (as displayed in the Object Browser)

Description: Model description text if entered in the setup tab.

Scenario: Name of the scenario (if applicable)

Retcode: Code indicating the status of the run convergence

LogLik: Loglikelihood

-2LL: –2 Loglikelihood

AIC: Akaike Information Criterion for each model run

BIC: Bayesian Information Criterion for each model run

nParm: Number of model parameters

nObs: Number of observations

nSub: Number of subjects

EpsShrinkage: Epsilon shrinkage

Posthoc Worksheet

Compares the Posthoc worksheet estimates of selected models. Model name, any description, scenario, replicate, ID, time, and one column for each structural parameter are listed.

Residuals Worksheet

This table stacks 'Residuals' worksheet results of each individual model with the addition of Model name and Description. It contains columns for:

Name: Model name

Description: Model description

Scenario: Scenario name if applicable

ID: Subject ID

Time: Time

TAD: Time after dose

PRED: Population prediction

IPRED: Individual prediction

DV: Observed dependent variable. This column header will display the name of the column mapped to CObs for each individual run. If models being compared have different column names mapped to CObs then as many columns as CObs names used will display.

IRES: Individual residual

PREDSE: Standard error of predicted value

Weight: Applied model weight

IWRES: Individual weighted residual

WRES: Weighted residual

CWRES: Corrected weighted residual

PCWRES: Predictive check weighted residuals

CdfPCWRES: Cumulative distribution function for predictive check weighted residual. This column would only have non-zero values if the PCWRES option was selected in the individual runs.

CdfDV: Cumulative distribution function for the dependent variable. This column would only have non-zero values if the PCWRES option was selected in the individual runs.

TADSeq: Time after dose sequence. This variable is used to produce plots with TAD.

ObsName: Observation Name. This variable distinguishes observations when there are multiple observed quantities.

Var. Inf. factor: The variance inflation factor (if applicable)

Theta Worksheet

This table stacks 'Theta' worksheet results of each individual model run with the addition of Model name and Description. It contains columns for:

Name: model name

Description: model description

Scenario: scenario name if applicable

Estimate: parameter estimates

Parameter: parameter name

Units: parameter units (if applicable)

Stderr: standard error

CV%: percent confidence of variation

lower % CI: lower confidence interval

upper % CI: upper confidence interval

Var. Inf. Factor: variance inflation factor (if applicable)

Comparison Results Worksheet

List all columns checked under options. Columns may include:

Hide: Models hidden

Compare: Models to be compared in the plot table

Method: phoenix model engine used for each model

Name: is the name of the object being compared

Description: optional description of the model

Lineage: List the 'parent' (i.e., reduced model) if the current model is a 'child' (derived from but with additional parameters) of another model

LogLik: estimate of the loglikelihood upon convergence

AIC: Akaike Information Criterion for each model run

BIC: Bayesian Information Criterion for each model run

-2(LL)Delta: Difference in -2LL for nested models only (Lineage-Child model)

AICDelta: Difference in AIC values. for nested models only (Lineage-Child model)

BICDelta: Difference in BIC values for nested models only (Lineage-Child model)

#params: Number of model parameters

#obs: Number of observations

#Subj: Number of subjects

p-value: Chi-Square p-value based on the Likelihood Ratio Test for nested models only.

Plot Table

Illustration of all the selected plots in rows for each model being compared (columns). This plot table is not editable, and it places side-by-side the plots from the individual runs for easy visualization. Any changes made to the individual run plots would be reflected in this Plot table. This table can be exported or printed upon right click.

Settings Text File

A text format file with the list of current columns, plots and models being compared.