Subscribe Bookmark RSS Feed

How to calculate reliability growth parameter standard errors?

dwaterson

Community Trekker

Joined:

Jul 25, 2014

I am looking to know the formula to calculate the standard errors for reliability growth parameters lamda and beta.

1 ACCEPTED SOLUTION

Accepted Solutions
Solution

The covariance matrix for the parameters is estimated using the Fisher information matrix. Although, using MS Excel, I cannot get the estimates to match exactly. They look to be the same up to 2 decimal points.

5 REPLIES
Solution

The covariance matrix for the parameters is estimated using the Fisher information matrix. Although, using MS Excel, I cannot get the estimates to match exactly. They look to be the same up to 2 decimal points.

michael_jmp

Staff

Joined:

Jun 23, 2011

This is described in the Statistical Details section of the Reliability Growth documentation:

Statistical Details for the Reliability Growth Platform

Michael Crotty
Sr Statistical Writer
JMP Development
dwaterson

Community Trekker

Joined:

Jul 25, 2014

Thank you, Michael. I did find the passage to which you are referring. I am actually looking for the math that is used, which is not in the provided link. Is there an article or book you can refer me to?

michael_jmp

Staff

Joined:

Jun 23, 2011

I think this paper might help:

Crow, L. H., (1982), “Confidence Interval Procedures for the Weibull Process with Applications to Reliability Growth,” Technometrics, Feb. 1982, 24:1, 67-72.

Michael Crotty
Sr Statistical Writer
JMP Development
dwaterson

Community Trekker

Joined:

Jul 25, 2014

Thank you, Michael. I will review this article.