Joint probability density: how to find the upper and lower bound in integration

probabilityprobability distributions

If the joint probability density of X and Y is given by

$$ f(x,y)= \begin{cases}
2(x^2-xy) && \text{ for } 0 <x<1,-x<y<x\\
0 && \text{elsewhere}\\
\end{cases} $$

and asking to find P(x+y<0.5)

as shown in the graph boundaries show be like this
graph

$$\int\limits_{0}^{1/2}\int\limits_{0}^{\frac{1}{2}-y}cxy^2(1-x)(1-y^2)\, dxdy = 1$$

but I am confuse because y depends on x, so it shouldn't be like this.

any explanation ?

thank you

Best Answer

You want to integrate over the intersection of $0<x<1$, $-x<y<x$, and $x+y<1/2$.

Because $y$ may be negative, then $x$ may be greater than $1/2$. Indeed when $x=1$, then $-1<y<-1/2$ is in the interval.

So the outer integration is over all of $x\in[0;1]$ and the inner is over $y\in[-x;\min\{x,1/2-x\}]$.

$$\int_0^1\int_{-x}^{\min\{x,1/2-x\}} 2(x^2-xy) \,\mathrm dy\,\mathrm dx$$

You can split this into a sum of two integrals, one over $x\in[0;1/2)$ and the second over $x\in[1/2;1]$.

Related Question