I am having difficulty understanding likelihood functions. If we have a probability density function of a random variable $X$ like this:

$f_{X} (x)=ax^2 + bx + c$ (i.e a simple quadratic polynomial), then how to go about finding the likelihood function for x.

All examples of a likelihood function involve some parameter theta. In this example, what is that theta?

  • 3
    $\begingroup$ No such $f$ can be a density function because its integral will diverge (or be zero), no matter what values $a$, $b$, or $c$ may have. You must specify a domain over which it is defined (with it implicitly equal to zero elsewhere). Neglecting this little detail has been the origin of many, many questions about likelihood functions here. I suspect you would benefit from searching our site for highly-voted questions about likelihoods; I know they provide plenty of the kinds of examples you seek. $\endgroup$
    – whuber
    Jun 17, 2015 at 19:21

1 Answer 1


In likelihood estimation we look for a function of the likelihood of the data, for various values of theta. So in MLE, the data is given, and theta is a variable. In your example, a, b and c are your theta's.

The likelihood function will be the product of the probability density function f(x) over all x (all the data). Your example will probably not have a solution. But, just for the idea, your likelihood function should represent the likelihood, given the data, for all possible theta values.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.