0% found this document useful (0 votes)
277 views1 page

Ee342 Final Cheat Sheet

The document discusses several probability distribution concepts: 1) It finds the CDF of a random variable X given its MGF. The CDF is defined in terms of an exponential and unit step function. 2) It calculates the probability mass function of a discrete random variable defined by a ceil function. 3) It discusses simulating draws from a continuous probability density function defined on the interval [0,1].

Uploaded by

Shane Takamune
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
277 views1 page

Ee342 Final Cheat Sheet

The document discusses several probability distribution concepts: 1) It finds the CDF of a random variable X given its MGF. The CDF is defined in terms of an exponential and unit step function. 2) It calculates the probability mass function of a discrete random variable defined by a ceil function. 3) It discusses simulating draws from a continuous probability density function defined on the interval [0,1].

Uploaded by

Shane Takamune
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

Consider a random variable X with MGF: MX(t)=2exp(-3t)/(2-t). Find the CDF.

Case Estimate MSE


Note that the exp(-3t) is a time shift of t+3. No ^
Therefore, the MGF reduces to: MX(t)=2/(2-t), which is an Exponential MGF. Observation X = mX σ X2
We now know the pmf is fX(x)=2exp(-2(t-3))u(t-3).Integrate fX(x) to obtain FX(x). ^
x
One E(Var(X |Y ))
Ans: F (x) =
∫ 2e−2(y−3)dy = (1− e−2(x +3) )u(x + 3) Observation X (Y ) = E(X |Y )
X
One
3
^ σ
X (Y ) = X2,Y (Y − m€ σ X2 (1− ρ 2 )
Y ) + mX
Consider the following Matlab command: x=ceil(4*rand(1,1000)-0.6). Find pX(i). Observation
The ceil command rounds up a number to the closest integer. €
Linear σY
This command gives numbers from –0.6 to 3.4, giving rounded values from 0 to 4. €
€ We have a number line from 0 to 4 where 0.6/4 of the numbers will be zero, €
and 0.4/4 of the numbers will be 4. The probability for 1, 2, & 3 are the same.
Ans: pX(i)= 0.15δ(i) + 0.25δ(i-1) + 0.25δ(i-2) + 0.25δ(i-3) + 0.1δ(I-4)  P(Y =€i | X = j)P(X = i) 
P(X = i | Y = j) =  
Write a Matlab code to simulate 1000 draws of pdf: fX(x)=2(exp(-x) – exp(-2x))u(x) €  P(Y = j) 
First, find CDF by integrating fX(x) in terms of y, replacing x with y and from 0x.
Ans: FX(x) = 2 - 2e-x – 1 + e-2x = u  u = (1- e-x)2  e-x =1-sqrt(u)  x = -ln(1-sqrt(u))

Joint pdf of X and Y is given by: fX,Y(x,y) = exp(-x-y)u(x)u(y).


a) Find the pdf of Z=min(X,Y) € Continuous Probability Distribution
FZ(z) = P(Z≤z) = P(min(X,Y)≤z) = P((X≤z) U (Y≤z)) pmf  fX(x) MGF  MX(t) Mean Variance
Uniform 1 a+b
= P(X≤z) + P(Y≤z) – P(X≤z, Y≤z) OR 1 – P(X>z, Y>z)
 1 – exp(-z)exp(-z) = 1- exp(-2z). Differentiate FZ(z) to obtain fZ(z) over (a,b) e tb − e ta (b − a) 2
Ans: fZ(z) = 2exp(-2z)u(z) b−a t(b − a) 2 12
b) Find the pdf of W=X/(X+Y) Exponential
Transformation Method λe− λx λ 1 1
Since we just have one variable, we can create another one to use.
Set V= X+Y so that….
λ−t λ λ2
€ €
σ 2 t 2  € mx
Normal
Jacobian = =v
 (x−m )2 
 − €2x   σ2
Gaussian 1
 fW,V(w,v) =vexp(-v)u(v) for 0≤w≤1
€ e 2σ  exp mx + 2 
Integrate fW,V(w,v) to obtain Marginal of W
2πσ
Ans: fW(w) = 1 for 0≤w≤1 € € €
OR € €
CDF Method Discrete Probability Distribution
FW(w) = P(W≤w) = P(X/(X+Y)≤w) = P(X≤w(X+Y)) = P(X(1-w)≤Yw) pmf  p(x) MGF  M (t) Mean Variance
=P(X((1-w)/w)≤Y)  € € X
np
Binomial  n x n−x ( pe t
+ 1− p) n np(1− p)
  p (1− p)
∞ ∞ dFW (w)  x
∫ ∫ exp(− y)dydx = w & fW (w) = =1
Ans: FW (w) = for 0≤w≤1 Poisson
0 x (1−w)
dw
λx exp{λ(e t −1)} λ λ
w e− λ €
x! € €
€ Geometric x−1 t 1 1− p
€ p(1− p) pe

€ 1− (1−€p)e t
€p p2

 €
 (x − mX )2 (y − mY )2 2ρ (x − mX )(y − mY ) 
Joint Gaussian  1 1
exp  + − 
2πσ X σY 1− ρ 2
2
 2(1− ρ )  σ X
2
σY2 € σ X σY 
€ € €
Example: You have 1000 free minutes/month on your cell. Assume all calls are
Central Limit Theorem: iid with pmf pX (k) = (0.25)(0.75) k−1,k = 1,2.... What is the max number of
X − mX i calls you can make such that there is a probability of 0.99 that you won’t go over
Normalize
€ RVs: W i = i , E(W i ) = 0 & Var(W i ) = 1 your minutes. Use CLT.
σXi Ans: Note that this is a Geometric pmf. Therefore, m X = 4 and σ X = 12

n z x2  x − m 1000 − 4 n   1000 − 4 n 
1 1 −
Let Z n = ∑
n i=1
W i  P(Z n ≤ z) =
2 π
∫ e dx = Φ(z) as n ∞ P = P(X ≤ 1000) = 0.99 ⇒ P 2x ≤
2

 = Φ
 12n


€ € −∞  x σ 12n 
Example:

Number of students enrolled in a course with Poisson RV mean 100. If there are 120 students or more, the class will be split in two. If there are €
less than 120, the class won’t
split. What is the probability the class will split?
Ans:
P(X ≥ 120) = P(X ≥ 119.5) = P
 X −100 119.5 −100  € Example: A user plays a game Xi times. Xi is a Poisson RV with λ1 = 0.15 .
≥  ≈ 1− Φ(1.95) ≈ 0.256
€ €  100 100  € The game is played 200 times. Later, a user plays an upgraded game Yj times.
Yj is a Poisson RV with λ2 = 0.25 . The game is also played 200 times.
Example:
If 10 fair dice are rolled, find the probability that the sum obtained is between 30 and 40. Find P(Y>X) =???
€ €
Ans: Let Xi denote the value of the i-th die, i=1,2,….10.
We calculate: E(X i ) = 3.5 and Var(X i ) = E(X i2 ) − E(X i ) 2 = 35 /12 Ans: The mean€of a Poisson RV is λ .
  P(Y > X ) = P(Y − X > 0) = P( Z > 0)
Using the Central Limit Theorem:  
29.5 − 35 X − 35 40.5 − 35  m − 0
P(29.5 ≤ X ≤ 40.5) = P ≤ ≤  by Guassian Approximation 
 350 350 350  Φ Z 
€    σZ 
€  12 12 12 
where m Z = mY €
− m X = 20 and σ Z = σ Y + σ X = 80
≈ 2Φ(1.0184) −1 ≈ 0.629 €
€ €

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy