Cumulative Distribution Function

Creator
Creator
Seonglae Cho
Created
Created
2023 Mar 7 2:33
Editor
Edited
Edited
2025 Apr 29 0:0

CDF

FX:tP(x<t)=tfX(u)duF_X: t \mapsto P(x \lt t) = \int_{-\infty } ^t f_X(u) du
For functions like min, you can first find the CDF using probability intervals and then differentiate it to get the PDF.
  • limx F(x)=0\lim_{x\rightarrow\ -\infty} F(x) = 0
  • limx F(x)=1\lim_{x\rightarrow\ \infty} F(x) = 1
  • F(x)F(x) is a non-decreasing function (
    Monotonic
    )
  • If two probability distributions have the same CDF, they’re the same
  • Two distinct probability distributions can’t have the same CDF, if two probability distributions have the same CDF, they’re the same. \mapsto
    Inverse transform sampling
FX(X)U(0,1)F_X(X) \sim \mathcal{U}(0,1)
Random Variable X follows the probability distribution, then input into FXF_X, results in a uniform
 
 

Proof that CDF is monotonic increase

Let t<tt \lt t', then FX(t)FX(t)0F_X(t') - F_X(t) \ge 0, since tt>0t' - t > 0 and fX0f_X \ge 0 since it is a
Probability Density Function
 
 
 
 
 

Recommendations