Release 11.0 Documentation for ANSYS
www.kxcad.net Home > CAE Index > ANSYS Index > Release 11.0 Documentation for ANSYS
PDCMAT, Rlab, Matrix, Name1, Name2, Corr, SLEVEL, Popt
Prints the correlation coefficient matrix.
MP ME ST PR PRN <> <> FL EM <> <> PP <>
Result set label. Identifies the result set to be used for postprocessing. A result set label can be the solution set label you defined in a PDEXE command (if you are directly postprocessing Monte Carlo Simulation results), or the response surface set label defined in an RSFIT command (for Response Surface Analyses).
Keyword for the type of correlation coefficient matrix.
IO | — | Matrix of correlation coefficients between random input variables and output parameters. |
II | — | Matrix of correlation coefficients between random input variables and other random input variables |
OO | — | Matrix of correlation coefficients between random output parameters and other random output parameters. |
S | — | Correlation coefficient between a single random parameter (input or output) and another random parameter (input or output). The probabilistic design parameters must be specified in Name1 and Name2 for this option. |
Parameter names. The parameters must have been previously defined as a random input variable or a random output parameter with the PDVAR command. These parameters are used for Matrix = S only and are ignored for the other Matrix keywords.
Keyword for the type of correlation coefficients to be used for the output.
RANK | — | Spearman rank-order correlation coefficient (default). |
LIN | — | Pearson linear correlation coefficient. |
Significance level. The value for the significance level must be between 0.0 and 1.0. The default value is 0.025 (2.5%).
Specifies whether the probabilities should be printed with the correlation coefficients.
0 | — | Print only the correlation coefficients. |
1 | — | Print both the correlation coefficients and the probabilities (default). |
Prints the correlation coefficient matrix.
If Rlab is left blank, then the result set label is inherited from the last PDEXE command (Slab), RSFIT command (RSlab), or the most recently used PDS postprocessing command where a result set label was explicitly specified.
For all correlation coefficients the probabilistic design tool evaluates the probability that the correlation coefficient can be neglected. The evaluation of this probability is based on statistical test theory. The larger this probability is the likelier it is that the correlation coefficient does not really reflect an observable statistical interdependence between the parameters involved. If this probability exceeds the significance level as specified by the SLEVEL parameter, the correlation coefficient should be regarded as negligible or insignificant. The higher the significance level SLEVEL, the more correlation coefficients are considered significant. Using the Popt parameter you can also get a list of the probabilities and review them as to how far they exceed the significance level or how far they stay below it.
The PDCMAT command cannot be used to postprocess the results in a solution set that is based on Response Surface Methods, only Monte Carlo Simulations.