The Diagnostics section of the Bayesian Estimates report in the Repeated Measures Degradation platform contains a plot of the observations and the fitted model, as well as options for other diagnostic plots. The Data and Fitted Model plot appears by default. It is important to verify that the fitted models can trace the data well. You can save the fitted values using the Save Fitted Values option in the Data and Fitted Model red triangle menu.
The Diagnostics red triangle menu contains the following options:
Residuals vs Time
Shows or hides a plot of the residuals for the fitted model versus the time variable. The Residuals vs Time red triangle menu contains an option that enables you to save the residuals to the data table.
Note: A violation of the identical and independently distribution assumption is generally not an issue for constructing the Life Distribution profilers.
Cox-Snell Residuals
Shows or hides a P-P plot of the Cox-Snell residuals, which can be used to validate the distributional assumption for the data. If the data points deviate far from the diagonal, then the distributional assumption might be violated. The Cox-Snell Residual P-P Plot red triangle menu contains an option that enables you to save the Cox-Snell residuals to the data table. See Meeker and Escobar (1998, sec. 17.6.1) for a discussion of Cox-Snell residuals.
Note: A violation of the distributional assumption is generally not an issue for constructing the Life Distribution profilers.
Plot Actual by Predicted
Shows or hides a plot of actual response values versus predicted response values.
Plot Residual by Predicted
Shows or hides a plot of residuals for the fitted model versus predicted response values.
Note: Many of the plots in the Diagnostics section have an option in the plot red triangle menu to emphasize a single value of the Label, System ID variable. The Hide to Deemphasize option hides all non-selected values of the Label, System ID variable. The Hide to Deemphasize option is selected by default.