Save Columns

Each Save Columns option adds one or more new columns to the current data table. Additional Save Columns options appear when the fitting method is REML. These are detailed in REML Save Columns Options.

Note the following:

•	When formulas are created, they are entered as Formula column properties.

•	For many of the new columns, a Notes column property is added describing the column and indicating that Fit Least Squares created it.

•

For the Predicted Formula and Predicted Values options, a Predicting column property is added. This property is used internally by JMP in conducting model comparisons (Analyze > Modeling > Model Comparison). When you fit many models, it is also useful to you because it documents the origin of the column.

Description of the Save Columns Options

Prediction Formula

Creates a new column called Pred Formula <colname> that contains both the formula and the predicted values. A Predicting column property is added, noting the source of the prediction.

Note: Pred Formula <colname> inherits certain properties from <colname>. These include Response Limits, Spec Limits, and Control Limits. If you change these properties for <colname> after saving Pred Formula <colname>, they will not update in Pred Formula <colname>.

See Prediction Formula for more details.

Predicted Values

Creates a new column called Predicted <colname> that contains the predicted values computed by the specified model. Both a Notes and a Predicting column property are added, noting the source of the prediction.

Note: Predicted <colname> inherits certain properties from <colname>. These include Response Limits, Spec Limits, and Control Limits. If you change these properties for <colname> after saving Predicted <colname>, they will not update in Predicted <colname>.

Residuals

Creates a new column called Residual <colname> that contains the observed response values minus their predicted values.

Mean Confidence Interval

Creates two new columns called Lower 95% Mean <colname> and Upper 95% Mean <colname>. These columns contain the lower and upper 95% confidence limits for the mean response.

Note: If you hold the SHIFT key while selecting the option, you are prompted to enter an α level for the computations.

Indiv Confidence Interval

Creates two new columns called Lower 95% Indiv <colname> and Upper 95% Indiv <colname>. These columns contain lower and upper 95% confidence limits for individual response values.

Note: If you hold the SHIFT key while selecting the option, you are prompted to enter an α level for the computations.

Studentized Residuals

Creates a new column called Studentized Resid <colname>, which contains the residuals divided by their standard errors.

Hats

Creates a new column called h <colname>. The column values are the diagonal values of the matrix

, sometimes called hat values.

Std Error of Predicted

Creates a new column called StdErr Pred <colname> that contains the standard errors of the predicted mean response.

Std Error of Residual

Creates a new column called StdErr Resid <colname> that contains the standard errors of the residual values.

Std Error of Individual

Creates a new column called StdErr Indiv <colname> that contains the standard errors of the individual predicted response values.

Effect Leverage Pairs

Creates a set of new columns that contain the X Leverage values and Y Leverage Residuals for each leverage plot. For each effect in the model, two columns are added. If the response column name is R and the effect is X, the new column names are:

•	X Leverage of X for R

•	Y Leverage of X for R

In the columns panel, these columns are organized in a columns group called Leverage.

Cook’s D Influence

Creates a new column called Cook’s D Influence <colname>, which contains values of the Cook’s D influence statistic.

StdErr Pred Formula

Creates a new column called PredSE <colname> that contains both the formula and the values for the standard error of the predicted values.

Note: The saved formula can be large. If you do not need the formula, use the Std Error of Predicted option.

Mean Confidence Limit Formula

Creates two new columns called Lower 95% Mean <colname> and Upper 95% Mean <colname>. These columns contain both the formulas and the values for lower and upper 95% confidence limits for the mean response.

Note: If you hold the SHIFT key while selecting the option, you are prompted to enter an α level for the computations.

Indiv Confidence Limit Formula

Creates two new columns called Lower 95% Indiv <colname> and Upper 95% Indiv <colname>. These columns contain both the formulas and the values for lower and upper 95% confidence limits for individual response values.

Note: If you hold the SHIFT key while selecting the option, you are prompted to enter an α level for the computations.

Save Coding Table

Creates a new data table whose first columns show the JMP coding for all model parameters. The last column gives the values of the response. If you entered more than one response column, all of these columns appear as the last columns in the coding table.

Note: The coding data table contains a table variable called Original Data that gives the name of the data table that was used for the analysis. In the case where a By variable is specified, the Original Data table variable gives the By variable and its level.

Prediction Formula

Pred Formula <colname> differs from Predicted <colname> in that it contains the prediction formula. Right-click in the Pred Formula <colname> column heading and select Formula to see the prediction formula. The prediction formula can require considerable space if the model is large. If you do not need the formula with the column of predicted values, use the Save Columns > Predicted Values option. For information about formulas, see the Using JMP book.

The Prediction Formula option is useful for predicting values in new rows or for use with the profilers. The profilers are available in the Fit Least Squares report (Factor Profiling). However, when your data table includes formula columns, you can also use the profilers provided in the Graph menu. When you are analyzing multiple responses, accessing the profilers from the Graph menu can be useful.

Note: If you select Graph > Profiler to access the profilers, first save the formula columns to the data table using Prediction Formula and StdErr Pred Formula. Then place both of these formulas into the Y, Prediction Formula role in the Profiler window. After you click OK, specify whether you want to use PredSE <colname> to construct confidence intervals for Pred Formula <colname>. Otherwise, JMP creates a separate profiler plot for PredSE <colname>.