Edit PDF Table

This operation allows you to edit a PDF table by changing the shape of the PDF graph. Because the PDF Editor uses a graphical approach for defining a PDF, the resulting data is, inevitably, somewhat imprecise. The most precise way to specify the PDF is through EMA code. PDFs created via EMA code can still be viewed graphically in the PDF Editor.

For more information about EMA, see Chapter 1 "External Model Access" on page 1.

Procedure 216: Editing a PDF Table

  1. Do one of the following:

    • Choose File > New…
    • File > Open… and select PDF from the pull-down menu.

    Result: The PDF Editor opens.

  2. Use the Edit Line Graph procedure (described in Edit Line Graph) to specify the line graph.

  3. Modify the graph with any of the following operations:

    • Use the Set Abscissa Lower Bound, Set Abscissa Upper Bound, Set Ordinate Lower Bound, and Set Abscissa Upper Bound operations (Set Abscissa and Ordinate Bounds) to set the upper and lower values for the PDF. By default, the empty line graph first displayed is set to have a domain of outcomes from 0 to 5 and a range of probability densities from 0 to 1.

      Note: Upper- and lower-bounds for the ordinate—the probability density—must be greater than or equal to zero. Also note that upper and lower bounds may be set either before or after a PDF is defined. If the bounds are set after a PDF is defined, the PDF does not change. However, the line graph is redrawn to match the new range of the ordinate and the new domain of the abscissa.

    • Use the Set Sampling Resolution operation (Set Sampling Resolution) to set the number of discrete points along the horizontal axis of the curve. The sampling resolution can be set either before or after a PDF is defined. The default PDF sample count is 500.

    • Use the Add Impulse operation (Add Impulse) to add one or more probability impulses to the PDF.

    • Use the Smooth operation (Smooth) to round out sharp edges in the function.

    • Use the Normalize operation (Normalize) to scale the entire range of probabilities, so that the overall probability for all events is one (1).