Analyze Variability - Storage
main topic     example     see also 

Stat > DOE > Factorial > Analyze Variability > Storage

You can store the fits and residuals, model information, and weights.  Minitab stores the checked values in the next available columns and names the columns.

Dialog box items

Fits (ln units): Check to store the fitted values for the natural log of the standard deviations.

Ratio Residuals: Check to store the ratio residuals.

Ln residuals: Check to store the log residuals.

Standardized ln residuals: Check to store the standardized log residuals.

Leverages: Check to store leverages.

Cook's distance: Check to store Cook's distance. This statistic is not available when the estimation method is maximum likelihood.

DFITS: Check to store the DFITS. This statistic is not available when the estimation method is maximum likelihood.

Coefficients: Check to store the coefficients, which are also displayed in the output. If Minitab removes some terms because the data cannot support them, the removed terms do not appear in the output.

Effects: Check to store the effects, which are also displayed in the output. Effects are not displayed or stored for the constant, covariates, or blocks.

Ratio Effects: Check to store the ratio effects.

Design matrix: Check to store the design matrix corresponding to your model.

When terms are removed because the data cannot support them, the design matrix does not contain the removed terms. The columns of the stored matrix match the coefficients that Minitab displays and stores.

Means: Check to store the least squares means that you specified in the Options subdialog.

Natural log transformation of standard deviation response: Check to store the transformed standard deviations.

Fits for original response: Check to store the back-transformed fitted standard deviations.

Means for original response: Check to store the back-transformed fitted means.

Weights for use in Analyze Factorial Design

Weights for replicates: Check to store weights based on the fitted variances for use in Analyze Factorial Design. The weights are the reciprocal of the fitted variances.

Weights for repeats: Check to store adjusted weights only if you have repeat measurements with some replicated points. Weights are the reciprocal of the adjusted variance.

Mean response: Enter the column containing the means of your repeats associated with the standard deviations. You can calculate and store the means in Pre-process responses.

Covariates (optional): Check to adjust for covariates and enter one or more columns containing the covariates.