why we need new lifetime model?

#1
Hello everybody!

as you see the title, I want to know that why we need new lifetime mode like exponential, weibull and other model which related to few model???

I want to know who was found these kind of model the first??

Thank you very much :):)
 
#3
Ummmm, Thanks for your reply!!

Actually i want to know why these model(like exponential,weibull or..) founded??(I know first of all used in medical(to evaluate number birth or dead of some special people))
But for example in engineering or in computer science! why we need foe example exponential lifetime??
 

hlsmith

Less is more. Stay pure. Stay poor.
#4
Have you google searched or used Wikipedia to read about how they were first surmised. Exponentiating has been around for awhile, while as you said Weibull can be in survival models (proportion of hazards regression).
 
#5
I searched on the google and also wikipedia but I want to know the case which is these distribution apply in for example computer science or mechanical engineering or ...

how hazard shape help them??
 

Mean Joe

TS Contributor
#6
Thanks for asking, it's been a while since I've done this. So I may be wrong...
But I think the exponential model implies a constant hazard rate.
And the others have non-constant hazard rate.

It's stuff like that, that people use to determine which model to employ.
 
#7
Thanks for your reply But,
I want to know the usage of (constant or non-constant) model!!!
why scientists like to use it??
 

Mean Joe

TS Contributor
#8
The non-constant model is often more realistic.
The constant model is simpler.

In my rusty opinion.

Some scientists like to use a couple of models. You can see if there's drastic changes in results from a couple differences in assumptions. Then you get an idea if you need to be real careful in your assumptions. And helps you pick the final model.