Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 4

Probability Density Function

A continuous random variable X on the domain a ≤ X ≤ b. The probability density function is a


d
function y=f ( x ) such that: P(c ≤ X ≤ d )= ∫ f (x)dx
c

Valid probability density function


y=f ( x ) is a valid probability density function of X , it must satisfy both:
(i) f (x) ≥ 0 for a ≤ X ≤ b
b
(ii) ∫ f (x )dx=1
a

Properties of Probability Density Function


For a continuous random variable X with pdf f (x) on the domain [a ; b]
(i) Mode is the value of x that maximizes f (x)
m
1
(ii) Median is the value m such that ∫ f ( x )dx=
a 2
(iii) Mean or Expected value (EX)
b
μ= EX=∫ x ⋅ f (x)dx
a

(iv) Variance
2
VarX =E X −¿
(v) Standard deviation
σ =√ VarX

Normal Distribution
Let X a continuous random variable with μ and standard deviation σ . Then X is
normally distributed if its probability density function y=f (x ):
Probability of normally distributed X
Use TI 84 to calculate the probability of normally distributed X ∼ N ( μ ; σ 2 )
● P( X =k )
nd
2 +Var →1 :normalpdf ¿

Input: x−value: k ; μ ;σ

● P( X ≤ k ); P( X ≥ k ); P(a ≤ X ≤ b)
nd
2 +Var →2 :normalcdf ¿

Input: lower :−1099 /k /a ;upper :k /1099 /b ; μ ; σ

Sampling Distribution Sn
Let X a continuous random variable with μ and standard deviation σ . Then the sum
of n independent observations of X is Sn= X 1 + X 2 +...+ X n is a random variable with
● Mean: μS =nμ n

● Standard deviation: σ s =√ n ⋅ σ
n

Distribution of sample mean X n


Let X a continuous random variable with μ and standard deviation σ . Then the mean
X + X +...+ X n
of n independent observations of X is X n= 1 2 is a random variable with
n
● Mean: μ X =μ n

σ
● Standard deviation: σ X = n
√n

Central Limit Theorem


Let X a random variable with μ and standard deviation σ .
(i) μS =nμ ; σ s =√ n ⋅σ
n n

σ
(ii) μ X =μ ; σ X =
n n
√n
(iii) If is NOT normally distributed, both X n and Sn are approximately normally
X
distributed if the sample size n is large enough (typically n ≥ 30)
(iv) If X itself is normally distributed, both X n and Sn are normally distributed for any n

Confidence Interval for Mean μ


Let X a random variable with unknown μ and known standard deviation σ . We want
to find a 95% confidence interval containing μ. We select a sample of size n which is
large enough and calculate the sample mean x .
Use TI 84 to calculate the 95% CI
STAT →TESTS →7 : Z−interval

Input: Stats
σ ; x ; n ; c−level :0.95
To calculate the 90% CI,
Invnorm > area 1- 90%/2, muy=0, xích ma= 1

Sample size n
If we are required to construct a confidence interval with a given width, we should
always choose the smallest n satisfying the requirement

σ
● 95% CI has width w=2 ×1.96 ×
√n
n
● As must be a whole number, then
n=¿
(integer part)

Example: if ¿
⇒ n=150+1=151

Claim about μ using CI [a ; b]


(i) Claim: μ=μ 0
● μ0 ∈CI ⇒ Do not reject claim
● μ0 ∉CI ⇒ Reject claim
(ii) Claim: μ ≤ μ 0
● μ0 >b ⇒ Do not reject claim
● μ0 <b ⇒ Reject claim

(iii) Claim: μ ≥ μ 0
● μ0 < a ⇒ Do not reject claim
● μ0 > a ⇒ Reject claim
SAT 4

You might also like