Home / DOE / Charts / Cube Plot / Options

Options¶

This page is the control by control reference for the dialog titled Create Cube Plots. What the finished chart shows is on Cube Plot, and how each corner value is formed is on Math Details.

This panel is not shared with any other chart. Every other chart in the family lets any factor take an axis. A cube corner is one end of a factor, so a cube axis needs a factor with exactly two positions and nothing else will do. That one rule shapes the whole dialog.

Outputs¶

The left side of the dialog is a box headed Outputs holding a tree with three levels: the worksheet, then each output on it, then each level of that output. Tick the levels you want a chart for. The dialog opens with every level on the active sheet already ticked.

The two upper levels are three-state. Ticking a worksheet or an output ticks everything under it, and clearing it clears everything under it. When only some of the children are ticked, the parent shows a partial state rather than a tick, so the tree tells you at a glance whether a selection is complete.

One worksheet at a time. Ticking anything on a second worksheet silently clears every tick on the one you had selected, and the options on the right rebuild for the new sheet. There is no warning and no way to draw charts from two source sheets in one Create.

If you press Create with nothing ticked, the dialog tells you At least one chart must be selected. and stays open.

Some charts can be built from either kind of sheet, and the two answer different questions.

A regression sheet gives the fitted model's prediction: a smooth value at any combination of settings, including combinations no run used. A design sheet gives the raw data: the average of the responses actually measured at each setting, and where a setting has replicates, their standard deviation. The design sheet can only speak about settings that were run.

Where both are offered, the dialog lists every eligible worksheet in the workbook and you pick one.

A binary or nominal output is offered from a regression sheet only. A corner on those outputs is a predicted probability, and the raw proportion of events at one corner is too coarse to stand in for it, so the design sheet is not offered for them. Asking for one anyway reports Cube Plot shows a binary or nominal output (a predicted probability) from a regression source only.

Under a quantitative output the two levels are the output's name followed by Y-Hat and S-Hat on a regression sheet, or Y-bar and S-bar on a design sheet, the second offered only when the output has a usable standard deviation model. Under a binary output the levels are written probability of followed by the event name and then the non-event name; under a nominal output there is one line per level and the last carries (reference level).

Cube axes¶

A box headed Cube axes holds three dropdowns, each listing only the factors that qualify as a cube axis. Horizontal axis (1) and Vertical axis (2) are required and default to the first two qualifying factors found. Depth axis, optional (3) adds a (None) choice at the top of its own list and defaults to (None), drawing a square with four corners; choosing a third factor there draws a cube with eight corners instead. A note under the three dropdowns states this: "Two axes draw a square with four corners; adding a depth axis draws a cube with eight."

The two or three chosen axes must all be different factors; pressing Create with the same factor chosen on more than one axis reports Each cube axis must be a different factor.

When the selected source has fewer than two factors that qualify as a cube axis, no square or cube can be formed. The Cube axes box is disabled and a message reads: "This source has fewer than two factors with exactly two levels or two observed values, so a cube cannot be drawn. Use a surface or contour plot instead."

What counts as a factor with exactly two positions is decided one way for each kind of factor:

A categorical factor qualifies when it has exactly two levels.
The block qualifies when the design has exactly two blocks. Its two corner labels are the block numbers 1 and 2.
A quantitative factor qualifies when the runs that were loaded hold exactly two distinct values of it. It is the data that decides, not the design: a factor run at three settings does not qualify even in a design that nominally has two levels, and one run at two settings qualifies even if it was declared continuous over a wider range. The lower of the two values is the low corner.

A design with a nested factor never gets a cube, whatever the axes. Asking for one reports Cube Plot is not available for a design with nested factors. and nothing is drawn. This is a whole-design refusal, not a rule about the factors you picked, so no choice of axes avoids it.

Three further messages exist for a factor that does not qualify, one naming its level count and one naming its count of distinct observed values. The dropdowns offer only qualifying factors, using the same test the engine applies, so a user cannot reach them from this dialog.

Set points¶

A box headed Set points (factors held constant) lists every factor that is not one of the chosen cube axes; a factor's value there is fixed at every corner, since a factor's role as an axis is what varies it corner to corner. Two separate things can hide this box, and either is enough on its own: it is hidden entirely for a design-sheet source, because a design-sheet cube averages the raw data over the off-cube factors and a held-constant set point has no effect there; and it is also hidden, on a regression-sheet source, whenever every factor of the design is one of the chosen cube axes, since there is then nothing left to hold.

Each row names one factor. A continuous factor's row is an editable box defaulting to the midpoint of the factor's uncoded valid range, with that range shown beside it (low to high). The box accepts only a number (rejecting anything else with Must be a number, checked when focus leaves the box) and carries no built-in minimum or maximum. A value typed outside the factor's range does not block the chart (regression source only; a design source ignores set points entirely): pressing Create instead raises a prompt naming the factor, the value typed, and its range, and asks whether to create the chart anyway. Answering no returns to the dialog with the value intact.

A categorical factor's row is a fixed dropdown of the factor's level names instead, defaulting to the last level shown.

A box never holds a value that is not a number. A box commits what you typed only when focus leaves it, and an entry that does not parse is rejected at that moment rather than committed. Pressing Create while a box is still showing a rejected entry reports One or more entries are not a valid number. Correct the highlighted boxes and try again., moves focus to the first such box, and creates nothing.

Before it draws¶

If anything you typed reaches outside the range the factor actually took in the data, the software asks before drawing. A set point is one source of that. On the dialogs that carry axis range boxes, which are Surface Plot, Contour Plot and Trellis, a From or to value counts too, as does a trellis slice range, so widening an axis past the experimental range raises the prompt on its own with every set point left alone. The prompt lists each offending entry with the factor's experimental range beside it, warns that the chart will extrapolate beyond the data, and asks whether to create it anyway. Answering no returns you to the dialog with your options intact.

This is a warning, not a refusal. A prediction outside the range of the data is an extrapolation of the fitted model, and the model was never tested there.

On a design source the prompt cannot appear. A design-sheet cube averages the runs it has and reads no set point at all, so there is no typed value that could reach outside a range.

While it draws¶

While charts are being written, a status line reports what is happening and a progress bar appears only when more than one chart is being drawn, since a bar for a single chart would sit at 100 percent from the moment it appeared.

Cancel stops the run cleanly between charts, not part way through one. A chart that has already been written stays; the sheet that was mid-write is deleted rather than left half finished. So cancelling never leaves a partial chart behind, but it can leave fewer sheets than you asked for.

Creating with the dialog left open lets you build several charts from one set of options without reopening it. The charts are identical either way: the only difference is whether the window closes afterwards.

When nothing is drawn¶

If the options you chose produce no drawable chart, the dialog reports There are no valid charts selected for the report. and writes nothing. Where the reason is specific, for example an axis rule or a source restriction, that more specific message is shown instead.

On this chart the specific messages do most of the work. A refusal that names its reason, the nested design or the binary output on a design sheet, is shown instead of the general message above.

The dialog is owned by the Excel window, so Excel cannot cover it while it is open. If it seems to have vanished, it is on another monitor rather than behind the workbook.