Scribd
Upload
5 views

Lecture 16

The lecture discusses independent normal random variables, detailing their properties and providing examples of calculations involving them. It covers the transformation of variables, the distribution of sums of independent normal variables, and applications in probability. Additionally, it includes examples related to polar transformations and the distribution of shots from a marksman.

Uploaded by

Thành Nguyễn
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
5 views

Lecture 16

The lecture discusses independent normal random variables, detailing their properties and providing examples of calculations involving them. It covers the transformation of variables, the distribution of sums of independent normal variables, and applications in probability. Additionally, it includes examples related to polar transformations and the distribution of shots from a marksman.

Uploaded by

Thành Nguyễn
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 7

Lecture 16: Independent Normal Random Variables

STOR 435, Spring, 2025

3/24/2025

435-Spring-2025 independent normal


Facts and Example 1

Pitman: Section 5.3


Facts:
If X = σZ + µ with parameters µ ∈ IR and σ > 0, then
Z ∼ N(0, 1) ⇐⇒ X ∼ N(µ, σ 2 ).
Let X1 , ..., Xn be independent with Xi ∼ N(µi , σi2 ), i = 1, ..., n.
Then Sn ≜ X1 + · · · + Xn ∼ N(µ1 + · · · µn , σ12 + · · · + σn2 ).
Example 1: Let X, Y, Z be iid N(0, 1) RV’s.

(a) P(2X − Y > 2Z + 1) = P(2X − Y − 2Z > 1)


= P(N(0, 9) > 1) = · · · = 0.3707.

435-Spring-2025 independent normal


Example 1 continued

(b) P(min{X, Y} > max{X, Y} − 2)


= P(X < Y, X > Y − 2) + P(X ≥ Y, Y > X − 2)
= P(−2 < X − Y < 0) + P(0 ≤ X − Y < 2)
= 2 P(0 ≤ N(0, 2) < 2) = · · · = 0.8414.

(c) SD(3X − 2Y + Z + 4)
q √
= 32 VarX + (−2)2 VarY + VarZ + Var4 = 14.

435-Spring-2025 independent normal


Example 2
Let X and Y be iid N(0, 1) RV’s. Define (R, Θ) via the polar
transformation x = r cos θ and y = r sin θ. Find the density
f(R,Θ) (r, θ). An “infinitesimal probability” argument could be
used. However, we show a standard procedure as follows.
First, r is the Jacobian determinant for the transformation
(r, θ) −→ (x, y). Next, for r > 0, θ ∈ [−π, π],
f(R,Θ) (r, θ) = r · f(X,Y) (r cos θ, r sin θ)
1 2 1 2 1 2 1
= r · √ e−(r cos θ) /2 √ e−(r sin θ) /2 = r e− 2 r .
2π 2π 2π
Note that
1 2
fR (r) = r e− 2 r , r > 0 (Rayleigh distribution);
and
1
, θ ∈ [−π, π].
fΘ (θ) =
2π
Hence R and Θ are independent.
435-Spring-2025 independent normal
Example 2 continued

Basic question: How to verify


Z ∞
1 2
I≜ √ e−z /2 dz = 1 ?
−∞ 2π

Answer: It follows from the polar transformation in Example 2


that
Z ∞Z ∞
2 1 −(x2 +y2 )/2
I = e dx dy
−∞ −∞ 2π
Z π Z ∞
1 2
= dθ · r e−r /2 dr
−π 2π 0
= 1.

435-Spring-2025 independent normal


Graphs for Example 3

435-Spring-2025 independent normal


Example 3

A marksman firing at a target produces a random scatter of


shots {(X, Y)} that follow roughly the same distribution as in
Example 2. Suppose approximately 50% of the shots fall in the
bull’s eye.
2
Following P(R ≤ r)√= 1 − e−r /2 = 1/2 yields the radius of
the bull’s eye r = 2 log 2. Moreover,
2
P(R ≤ 2r) = 1 − e−(2r) /2 = 15/16.
Note that the distribution of (X, Y) is rotation-invariant. For
the square A with four vertices (r, 0), (0, r), (−r, 0), (0, −r),
√ √
P((X, Y) ∈ A) = P(|X| ≤ 2r/2, |Y| ≤ 2r/2)
p
= [P(|X| ≤ log 2)]2 = 0.3521.

435-Spring-2025 independent normal

You might also like