Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Does anyone know, is $h\left(z,w\right):=\frac{\left|zw\right|}{\left|z\right|+\left|w\right|}$ for $z$ and $w$ in the unit disk $\mathbb{D}$ of the complex plane a (pluri)subharmonic function? Thanks!

share|cite|improve this question

I didn't see an easy non-computational solution, but write $$h(z,w) = \frac{\sqrt{z\bar z}\sqrt{w\bar w}}{\sqrt{z \bar z}+\sqrt{w \bar w}}$$ and start computing. (I admit using Maple, since I'm lazy). We get $$\frac{\partial^2 h}{\partial z \partial\bar z} = \frac14\frac{|z|^2\big(|w|-|z|\big)}{|w|\,\big(|z|+|w|\big)^3},$$ $$\frac{\partial^2 h}{\partial w \partial\bar w} = \frac14\frac{|w|^2\big(|z|-|w|\big)}{|z|\,\big(|z|+|w|\big)^3}$$ and $$\frac{\partial^2 h}{\partial z \partial\bar w} = \frac12\frac{z \bar w}{\big(|z|+|w|\big)^3}.$$

Hence (after lots of tedious algebra) $$\Delta h = 4\left( \frac{\partial^2 h}{\partial z \partial\bar z} + \frac{\partial^2 h}{\partial w \partial\bar w} \right) = \frac{\big(|z|-|w|\big)^2(|z|^2+|zw|+|w|^2)}{|zw|\big(|z|+|w|\big)^3} \ge 0$$ so $h$ is subharmonic (outside the coordinate axis, and it's easy to check that it's subharmonic even there). On the other hand $$\frac{\partial^2 h}{\partial z \partial\bar z}\frac{\partial^2 h}{\partial w \partial\bar w} - \frac{\partial^2 h}{\partial z \partial\bar w}\frac{\partial^2 h}{\partial w \partial\bar z} = -\frac{1}{16} \frac{|zw|}{(|z|+|w|)^4}$$ which is negative for most points in the bidisc. This shows that $h$ is not plurisubharmonic.

share|cite|improve this answer

I don't have an easier proof of subharmonicity than the one by mrf, but here is the non-psh part. Note that $h$ is polyradial: it depends only on $|z|$ and $|w|$. Such a function is plurisubharmonic (on a Reinhardt region where $zw\ne 0$) if and only if $h(e^x,e^y)$ is a convex function of $(x,y)$: see Proposition 1.14.40 here. Looking at the line $y=-x$, we get $h(e^x,e^{-x})=1/(2\cosh x)$ which is certainly not convex near $0$, having strict maximum at $0$. Well, the line $y=-x$ does not actually meet the region $\{x<0,y<0\}$ which we are to consider; but since convexity is preserved by pointwise limits, it must fail on the line $y=-\epsilon-x$ for all small $\epsilon>0$.

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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