fstats_experimental_design Module

Model degrees of freedom.

Residual degrees of freedom.

Total degrees of freedom.

F-statistic.

Model mean square.

Residual mean square.

P-value for the F-test.

R-squared value.

Model sum of squares.

Residual sum of squares.

Total sum of squares.

Interpretation of comparison results.

Difference in degrees of freedom.

Degrees of freedom for full model.

Degrees of freedom for reduced model.

F-statistic for model comparison.

P-value for the F-test.

Residual sum of squares for full model.

Residual sum of squares for reduced model.

True if models differ significantly (p < 0.05).

P-value for the overall F-statistic.

Overall F-statistic for the model.

Mean of the response variable.

Number of observations.

Number of model parameters (including intercept).

The coefficient of determination (R²), range [0, 1].

The adjusted R² accounting for model complexity.

Residual standard error (standard deviation of residuals).

Root mean square error.

Higher is better.

Range [0,1]. Higher is better (max=1 for orthogonal designs).

G-efficiency: 1 - (max_prediction_variance / avg_prediction_variance)

True if design is perfectly orthogonal.

Number of factors.

Number of model parameters.

Number of design runs.

Orthogonality measure: $1.0$ if perfectly orthogonal, $<1.0$ otherwise.

The model coefficients, where N is the number of factors. The coefficients are ordered as follows, with each group included only if the model has a sufficient number of interaction levels.

The unscaled parameter covariance matrix $\left( X^{T} X \right)^{-1}$, where $X$ is the design matrix used to fit the model. Only the parameters included in the model (map = true) are represented, in coefficient order.
This is populated by doe_fit_model, and is used to compute prediction intervals.

The residual degrees of freedom of the fit, $m - p$. A value of zero indicates the fit information is not available (e.g. the model was not created by doe_fit_model).

An array denoting if a model coefficient should be included as part of the model (true), or neglected (false).

The residual mean squared error of the fit, $SS_{res} / (m - p)$. This is only meaningful if dof > 0.

The number of interaction levels.

Statistical information for each model parameter.

Whether optimization converged.

Tolerance used for convergence.

Number of iterations to converge.

Optimization method used.

Optimal factor settings (in coded scale).

Optimal factor settings (in natural scale).

Predicted response at optimal point.

Confidence level (e.g., 0.95 for 95% CI).

Lower confidence interval bounds.

Upper confidence interval bounds.

Predicted response values.

Lower prediction interval bounds.

Upper prediction interval bounds.

Observed response values.

Predicted response values.

Mean of residuals (should be ~0).

Standard deviation of residuals.

Raw residuals: y - y_predicted.

Standardized residuals for outlier detection.

Base fitted model.

Interaction coefficients.

Model intercept.

Linear coefficients for each factor.

Number of factors.

Quadratic coefficients (main effects squared).

Predicted response at design center (coded 0,0,...,0).