Joint probability function

Definition (Joint probability mass function)

For 2 RVs: Let X,YX,Y be two discrete RVs defined on the same sample space SS of a random experiment and taking values in the sets X,Y\mathscr{X},\mathscr{Y}. Then the joint pmf of XX and YY is px,y(x,y):=P(X=x,Y=x) , xX, yYp_{x,y}(x,y):=P(X=x,Y=x) \ , \ x\in\mathscr{X}, \ y\in\mathscr{Y}

Proposition (Properties)

  1. p(x,y)0 xX,yYp(x,y)\ge 0 \ \forall x\in\mathscr{X},y\in\mathscr{Y}
  2. p(x,y)=0 xX,yYp(x,y)=0 \ \forall x\notin\mathscr{X},y\notin\mathscr{Y}
  3. xXyYp(x,y)=1\sum_{x\in\mathscr{X}}\sum_{y\in\mathscr{Y}}p(x,y)=1
  4. For AX×YA\subset\mathscr{X}\times\mathscr{Y}, P((X,Y)A)=(x,y)Ap(x,y)P((X,Y)\in A)=\sum_{(x,y)\in A}p(x,y)

Definition (Joint probability density function)

XX and YY are jointly continuous if there exists a non-negative function f:R×R[0,)f:\mathbb{R}\times\mathbb{R}\to[0,\infty) such that for any reasonable set CR2C\subset\mathbb{R}^2 (measurable), we have P((X,Y)C)=Cf(x,y)dxdyP((X,Y)\in C)=\int\int_Cf(x,y)dxdyRVs XX and YY are called jointly continuous and ff is their joint pdf.

Linked from