Re: probability vs. likelihood

Quoting Wikipedia: "probability" allows us to predict unknown outcomes
based on known parameters, then "likelihood" allows us to estimate
unknown parameters based on known outcomes.
With probability, you calculate how well the observed data fits a
given set of parameters of a distribution, with likelihood you
calculate how well a supposed set of parameters of a distribution fit
the observed data.
The use of one or another depends on what you want to study with your
data, for instance, with risk prediction, if politician A says that
the risk of X has decreased, your null hypothesis is that you data
came from a distribution where risk X at time 0 = risk X at time 1,
and the data should be very unlikely (<5%) to have come from such a
distribution for the politician to say this (i.e. risk t0 > risk t1).
With risk likelihood, you can calculate the most likely distribution,
given the data, and calculate that with 95% confidence the underlying
distribution, and if all possible distributions have risk t0 > risk
t1, the politician is also right.
The argument for using likelihood usually is because you don't have to
assume any underlying distribution parameters (i.e. how much risk
reduction, if any), since that is what you are calculating, but the
statistical significance of both tests will be the same.
Hope this helps.
Keo.

On Aug 10, 11:12 am, khacker <kenneth.hac...@xxxxxxxxx> wrote:
I recently heard some government analysts argue that risk prediction
should be based more on likelihood than on probability. I never heard
this distinction before and wonder what primer I can read about the
differences.

Thanks,

Ken Hacker
Dept. of Communication Studies
New Mexico State University

.

Relevant Pages

  • Re: Maximum Likelihood Estimation
    ... In my understanding, having the ... likelihood estimation procedure seeks to find the set ... distribution say N. ...
    (sci.stat.math)
  • Re: Median likelihood?
    ... of the likelihood function or the median of the log likelihood ... Median is of a probability distribution (such as the sampling ... distribution is equal to the parameter being estimated, an estimator ...
    (sci.stat.math)
  • Re: Median likelihood?
    ... of the likelihood function or the median of the log likelihood ... Median is of a probability distribution (such as the sampling ... distribution is equal to the parameter being estimated, an estimator ...
    (sci.stat.math)
  • Re: How to combine the results of two Maximum Likelihood Estimation?
    ... inappropriate answer to the question Lucy ASKED. ... What you said presumes that the joint likelihood function of the ... the entire joint distribution should have been ...
    (sci.stat.math)
  • Re: Top 10 Text Values
    ... >>header, which in all likelihood would be the case, and wouldn't affect ... While I usually try to eliminate any risk, the risk here is so remote ... With regards to the two solutions, I was surprised to see that there ...
    (microsoft.public.excel.worksheet.functions)