1
$\begingroup$

I have a probability density function, $f_{X,Y}(x,y)$. I can calculate moments of $f$ easily, namely, I can calculate $\int\int x^ay^bf_{X,Y}(x,y)\,dx\,dy$ over the range of $X$ and $Y$, for any particular $a$ and $b$ I like.

If I calculate first and second moments, I could approximate $f$ with a gaussian. Suppose I calculated the third and/or fourth moments too. What might I approximate $f$ by, taking into account these extra moments?

[NB: if it helps, I know that $(X,Y)$ are bound within a triangle (say, $X\geq 0$, $Y\geq 0$, $X+Y\leq 1$, but it's up to me). I also expect that $f$ has the form of an algebraic function of $x$ and $y$, but I'm happy to ignore this]

$\endgroup$
5
  • 1
    $\begingroup$ Calculate the skewness and excess kurtoses and then depending on the results choose likely distributions. For instance if skewness=0, or at least more or less zero, then this eliminates many possibilities. $\endgroup$
    – user121049
    Sep 19, 2017 at 6:44
  • $\begingroup$ You may find this useful $\endgroup$
    – caverac
    Sep 19, 2017 at 7:37
  • $\begingroup$ @caverac did you just link back to my own question? $\endgroup$
    – Michael Hartley
    Sep 21, 2017 at 12:36
  • $\begingroup$ :) Sorry, for some reason I copied the wrong link $\endgroup$
    – caverac
    Sep 21, 2017 at 12:39
  • $\begingroup$ This looks like a two-dimensional moment problem. $\endgroup$
    – Keith McClary
    Oct 22, 2018 at 0:11

0

Reset to default

You must log in to answer this question.

Browse other questions tagged .