True 3P MLE method

[S. Nath]

1. What is this method?
2. How is this different from the regression method of estimating the 3rd weibull parameter?

Thanks

[RS Support]

If not selected Weibull++ first determines the estimate of the location parameter, by nonlinear regression. Once this estimate is determined, Weibull++ uses MLE to estimate the other two parameters. This methodology is generally more appropriate for the analysis of life data.

With the Use True 3-P MLE on Weibull option selected, the results are purely MLE on all three parameters. However, a pure MLE solution on a 3P Weibull is problematic in certain regions of beta yielding results that may not make sense in the context of life data analysis (e.g. very large negative values for gamma etc.)

[S. Nath]

In order to alter parameters is it required to select the true 3P MLE option? Or can it be done without selecting this option? How?
The reason for these questions is as follows,
With Weibull 5.32 I was able to alter parameter without any problem, but with Weibull 6 its not so.
thanks in advance.

[Pantelis]

Nath,

That seems to be the case. Let me find out why it was changed from 5 and will get back to you in this forum.

[S. Nath]

Pantelis,
We frequently use the Reliasoft's 3 parameter weibull since it models our data adequately, and it quantifies a finite failure-free (durability) life. Maybe there is no difference between weibull 5 and 6 wrt this feature, but I just want to make sure, thats all. And I want to continue using the alter parameter feature with the regression option.
Thanks.