Relationship To Poisson Models
There is a relationship between proportional hazards models and Poisson regression models which is sometimes used to fit approximate proportional hazards models in software for Poisson regression. The usual reason for doing this is that calculation is much quicker. This was more important in the days of slower computers but can still be useful for particularly large data sets or complex problems. Authors giving the mathematical details include Laird and Olivier (1981), who remark
"Note that we do not assume is true, but simply use it as a device for deriving the likelihood."
The book on generalized linear models by McCullagh and Nelder has a chapter on converting proportional hazards models to generalized linear models.
Read more about this topic: Proportional Hazards Models
Famous quotes containing the words relationship to, relationship and/or models:
“Sometimes in our relationship to another human being the proper balance of friendship is restored when we put a few grains of impropriety onto our own side of the scale.”
—Friedrich Nietzsche (1844–1900)
“It would be a fallacy to deduce that the slow writer necessarily comes up with superior work. There seems to be scant relationship between prolificness and quality.”
—Fannie Hurst (1889–1968)
“French rhetorical models are too narrow for the English tradition. Most pernicious of French imports is the notion that there is no person behind a text. Is there anything more affected, aggressive, and relentlessly concrete than a Parisan intellectual behind his/her turgid text? The Parisian is a provincial when he pretends to speak for the universe.”
—Camille Paglia (b. 1947)