Launch the Fit Life by X Platform | JMP 13.2

JMP Support 800.450.0135 (US)

Documentation Feedback
Your feedback is important to us. us any comments about our documentation.

Reliability and Survival Methods • Fit Life by X • Launch the Fit Life by X Platform

•

Launch the Fit Life by X Platform

Launch the Fit Life by X platform by selecting Analyze > Reliability and Survival > Fit Life by X.

The Fit Life by X Launch Window

The Fit Life by X launch window contains the following options:

Y, Time to Event

Identifies the time to event (such as the time to failure) or time to censoring. With interval censoring, specify two Y variables, where one Y variable gives the lower limit and the other Y variable gives the upper limit for each unit. For details about censoring, see Event Plot in Life Distribution.

Identifies the accelerating factor.

Identifies censored observations. Select the value that identifies right-censored observations from the Censor Code menu beneath the Select Columns list. The Censor column is used only when one Y is entered.

Identifies frequencies or observation counts when there are multiple units. If the value is 0 or a positive integer, then the value represents the frequencies or counts of observations for each row when there are multiple units recorded.

Identifies a column that creates a report consisting of separate analyses for each level of the variable.

After selecting the Censor column, select the value that designates right-censored observations from the list. Missing values are excluded from the analysis. JMP attempts to detect the censor code and display it in the list.

Determines the relationships between the event and the accelerating factor. Examples include transformation using the following acceleration models:

–

Arrhenius Celsius: μ = b0 + b1 * 11605 / (X+273.15)

–

Arrhenius Fahrenheit: μ = b0 + b1 * 11605 / ((X + 459.67)/1.8)

–

Arrhenius Kelvin: μ = b0 + b1 * 11605/X

–

Voltage: μ = b0 + b1 * log (X)

–

Linear: μ=b0 + b1* X

–

Log: μ = b0 + b1 * log(X)

–

Logit: μ = b0 + b1 * log(X/(1-X))

–

Reciprocal: μ = b0 + b1/X

–

Square Root: μ = b0 + b1 * sqrt(X)

–

Box-Cox: μ = b0 + b1 * Boxcox(X)

–

Location: means μ is different for every level of X.

–

Location and Scale: means μ and σ are both different for every level of X (equivalent to a Life Distribution fit with X as a By variable).

If you select Location or Location and Scale, a message might appear stating that “Analysis must exclude groups that are all right censored to continue Nested Model Tests. Do you want to continue?” Click Yes to continue the analysis. Right censored groups indicate groups where all observations are right censored. The message does not appear if your sample data does not include right censored observations.

See Custom Relationship if you want to use a Custom relationship for your model.

Nested Model Tests

Appends a nonparametric overlay plot, nested model tests, and a multiple probability plot to the report window.

Enables you to enter a baseline value for the explanatory variable, X, of the acceleration factor. You can also set the baseline after launching the platform using the Set Time Acceleration Baseline option from the Fit Life by X red triangle menu.

Specifies one distribution (Weibull, Lognormal, Loglogistic, Fréchet, SEV, Normal, Logistic, LEV, or Exponential distributions) at a time. Lognormal is the default setting.

Select Confidence Interval Method

Displays the method for computing confidence intervals for the parameters. The default is Wald, but you can select Likelihood instead. The Wald method is an approximation and runs faster. The Likelihood method provides more precise parameters but takes longer to compute.

Note: The Confidence Interval Method preference enables you to select the Likelihood method as the default confidence interval method. You can change this preference in Preferences > Platforms > Fit Life by X.

•

Help created on 9/19/2017