Hi All, I am trying to predict hazard rate and MTTF with Eyring model (see attached file). Using Parametric Survival platform. In Parameter Estimates report there is scale parameter "sigma" but no separate location parameters for Loglogistic distribution. Only graphic representation of separate location and separate location & scale are given in Parametric Survival platform. Is there way to get values of separate location parameters for chosen probabilistic distribution? Example8.9.jmp

nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

Separate Location parameters using Parametric Survival platform

ptolomey

Hi All,

I am trying to predict hazard rate and MTTF with Eyring model (see attached file).

Screenshot from 2018-11-13 20-26-02.png

Using Parametric Survival platform.

Parametric Survival Dialog.jpg

In Parameter Estimates report there is scale parameter "sigma" but no separate location parameters for Loglogistic distribution.

Parameters Estimates.jpg

Only graphic representation of separate location and separate location & scale are given in Parametric Survival platform.

Graphic Representation.jpg

Is there way to get values of separate location parameters for chosen probabilistic distribution?

Example8.9.jmp

Accepted answers

Mark_Bailey

So, use the parameter estimates to compute the Location parameter for each of the conditions:

Location = -7.2947 + 0.533 * Arrhenius(Temp) + -1.11 * Log(Voltage)

All comments

Mark_Bailey

The location is not constant like the scale. The location parameter is the back transformed function of the linear predictor.

ptolomey

Hi Mark.

Thank you for reply.

I made same analysis in Minitab Accelerated Life Testing plaform.

In the report there is Location parameter for each Temperature-Voltage pair.

By the way vaue of Scale parameter in Minitab is mutches value of same parameter in JMP 0.34

Report Graph.jpg

Mark_Bailey

So, use the parameter estimates to compute the Location parameter for each of the conditions:

Location = -7.2947 + 0.533 * Arrhenius(Temp) + -1.11 * Log(Voltage)

ptolomey

Hi Mark,

You absolutely right.

For standardized Loglogistic distribution: Location = Ln(T50)

Thank you.