Properties of the Joint CDF of Two Continuous Random Variables

Use the joint CDF FX,Y(x,y)=P(Xx,Yy)F_{X,Y}(x,y)=P(X\le x, Y\le y) to recover marginal CDFs and to compute probabilities of rectangular events via an inclusion–exclusion formula.

Step 1 of 119%

Tutorial

Basic Properties and Marginal CDFs

The joint CDF of two continuous random variables XX and YY is

FX,Y(x,y)=P(Xx,  Yy).F_{X,Y}(x,y) = P(X \le x,\; Y \le y).

The joint CDF satisfies four basic properties:

  1. Limits at -\infty: limxFX,Y(x,y)=0\lim\limits_{x\to -\infty} F_{X,Y}(x,y) = 0 for every y,y, and limyFX,Y(x,y)=0\lim\limits_{y\to -\infty} F_{X,Y}(x,y) = 0 for every x.x.
  2. Limit at ++\infty: limx,yFX,Y(x,y)=1.\lim\limits_{x,y\to \infty} F_{X,Y}(x,y) = 1.
  3. Monotonicity: FX,YF_{X,Y} is non-decreasing in each argument.
  4. Marginal CDFs: The CDFs of XX and YY alone are recovered by sending the other variable to +:+\infty{:}
FX(x)=limyFX,Y(x,y),FY(y)=limxFX,Y(x,y).F_X(x) = \lim_{y\to\infty} F_{X,Y}(x,y), \qquad F_Y(y) = \lim_{x\to\infty} F_{X,Y}(x,y).

For instance, suppose

FX,Y(x,y)=(1ex)(1ey)for x,y0.F_{X,Y}(x,y) = (1-e^{-x})(1-e^{-y}) \quad \text{for } x,y \ge 0.

Sending y,y\to\infty, the factor 1ey1,1-e^{-y} \to 1, so the marginal CDF of XX is

FX(x)=(1ex)1=1ex,x0.F_X(x) = (1-e^{-x})\cdot 1 = 1 - e^{-x}, \qquad x\ge 0.

Sending both x,yx,y\to\infty gives FX,Y(,)=11=1,F_{X,Y}(\infty,\infty) = 1\cdot 1 = 1, confirming property 2.

navigate · Enter open · Esc close · ⌘K/Ctrl K toggle