FORMULAS 𝐸(𝑋𝑌) = 𝐸(𝑋)𝐸(𝑌) *if independent random

Chapter 1: variables
A priori: 𝑃(𝐴) =
𝑛 *Properties of the Variance
𝑁
2
𝑛! 𝑉𝑎𝑟(𝑎𝑋) = 𝑎 𝑉𝑎𝑟(𝑋)
Permutation: 𝑛𝑃𝑟 = (𝑛−𝑟)!
𝑉𝑎𝑟(𝑏) = 0
𝑛!
Combination: 𝑛𝐶𝑟 = 2
𝑟!(𝑛−𝑟)! 𝑉𝑎𝑟(𝑎𝑋 + 𝑏) = 𝑎 𝑉𝑎𝑟(𝑋)
Properties of Probability Function 𝑉𝑎𝑟(𝑋 + 𝑌) = 𝑉𝑎𝑟(𝑋) + 𝑉𝑎𝑟(𝑌)
𝑐
𝑃(𝐴 ) = 1 − 𝑃(𝐴) 𝑉𝑎𝑟(𝑋 − 𝑌) = 𝑉𝑎𝑟(𝑋) + 𝑉𝑎𝑟(𝑌)
𝑐 2 2
𝑃(𝐴 ∩ 𝐵 ) = 𝑃(𝐴) − 𝑃(𝐴 ∩ 𝐵) 𝑉𝑎𝑟(𝑋) = 𝐸(𝑋 ) − [𝐸(𝑋)]
𝑃(𝐴 ∪ 𝐵) = 𝑃(𝐴) + 𝑃(𝐵) − 𝑃(𝐴 ∩ 𝐵)
𝑐
𝑃((𝐴 ∩ 𝐵) ) = 1 − 𝑃(𝐴 ∩ 𝐵) *Binomial Distribution
𝑛 𝑜𝑓 𝑠𝑢𝑐𝑐𝑒𝑠𝑠
𝑐 Random variable:
𝑃((𝐴 ∪ 𝐵) ) = 1 − 𝑃(𝐴 ∪ 𝐵) 𝑁 𝑜𝑓 𝑡𝑟𝑖𝑎𝑙𝑠
If 𝑋~𝐵𝑖𝑛𝑜𝑚𝑖𝑎𝑙(𝑛, 𝑝), where 𝑛 is the number
*Independents events if and only if: of trials, 𝑝is the probability of success:
𝑃(𝐴 ∩ 𝐵) = 𝑃(𝐴) × 𝑃(𝐵) 𝐸(𝑋) = 𝑛𝑝 and 𝑉𝑎𝑟(𝑋) = 𝑛𝑝(1 − 𝑝)
CDF: 𝐹(𝑥) = 𝑃(𝑋 ≤ 𝑥)
PMF: 𝑃(𝑋 = 𝑥) *Normal Distribution
PDF: 𝑃(𝑋 ≤ 𝑐) Normally distributed if the PDF is given by:
1 𝑋−µ 2
*Notes about the CDF: 𝑓(𝑥) =
1
𝑒
−2( σ
)

𝑃(𝑋 ≤ 𝑏) − 𝑃(𝑋 ≤ 𝑎) σ 2π
2
= 𝐹(𝑏) − 𝐹(𝑎) 𝑋~𝑁𝑜𝑟𝑚𝑎𝑙(µ, σ )
*Expected value of X or µ𝑥 Φ(•)𝑜𝑟 𝐹𝑍(•) - CDF of a standard normal
𝑛
= ∑ 𝑥𝑖𝑓(𝑥𝑖) random variable
𝑖=1 𝑃(𝑍 ≤ 𝑎) = 𝑃(𝑍 < 𝑎) = Φ(𝑎)
E(X) as a Weighted Mean 𝑃(𝑍 > 𝑎) = 𝑃(𝑍 ≥ 𝑎) = 1 − Φ(𝑎)
𝑛
∑ 𝑊𝑖𝑋𝑖 𝑃(𝑎 < 𝑍 < 𝑏) = 𝑃(𝑎 ≤ 𝑍 ≤ 𝑏) = 𝑃(𝑎 ≤ 𝑍 < 𝑏)
𝑋𝑤 = 𝑖=1
𝑛
𝑃(𝑎 < 𝑍 ≤ 𝑏) = Φ(𝑏) − Φ(𝑎)
∑ 𝑊𝑖
𝑖=1
2 2 2
*Variance of X 𝑋~𝑁(µ, σ ) → 𝑌~𝑁(𝑎µ + 𝑏, 𝑎 σ )
2 2 2 2 𝑋−µ 𝑎−µ 𝑎−µ
σ = 𝑉𝑎𝑟(𝑋) = 𝐸(𝑋 − µ) = 𝐸(𝑋 ) − [𝐸(𝑋)] 𝑃(𝑋 ≤ 𝑎) = 𝑃( σ
≤ σ
) = 𝑝(𝑧 ≤ σ
)
2 2 2
Proof: 𝐸(𝑋 − µ) = 𝐸[𝑥 − 2µ𝑥 + µ ]
2 2 𝑃(𝑍 > 𝑧α) = α is equivalent to:
= 𝐸(𝑥 ) − 2µ𝐸𝑥 + µ
= 𝐸(𝑥 ) − µ
2 2 𝑃(𝑍 ≤ 𝑧α) = 1 − α
*Properties of the Mean
𝐸(𝑋 − µ) = 0
𝐸(𝑎𝑋) = 𝑎𝐸(𝑋)
𝐸(𝑏) = 𝑏
𝐸(𝑋 + 𝑌) = 𝐸(𝑋) + 𝐸(𝑌)
𝐸(𝑋 − 𝑌) = 𝐸(𝑋) − 𝐸(𝑌)
 𝐸(𝑋1)+𝐸(𝑋2)+...𝐸(𝑋𝑛)
= 𝑛
𝑛 𝑛
Chapter 2: ∑ 𝐸(𝑋𝑖) ∑µ
𝑖=1 𝑖=1 𝑛µ
2 2
= 𝑛
= 𝑛
= 𝑛 = µ
𝑆. 𝐸. (𝑋) = 𝑉𝑎𝑟( (𝑋) = 𝐸( (𝑋 ) − [𝐸( (𝑋)] 𝑋1+𝑋2+....𝑋𝑛
Var (𝑋) = 𝑉𝑎𝑟 ( 𝑛
)
Sampling Distribution of 𝑋 𝑉𝑎𝑟(𝑋1)+𝑉𝑎𝑟(𝑋2)+...𝑉𝑎𝑟(.𝑋𝑛)
= 2 )
𝑛
𝑛 𝑛
2
∑ 𝑉𝑎𝑟(𝑋𝑖) ∑σ
Theorem Theorem 𝑖=1 𝑖=1 𝑛σ
2
σ
2

11.1 11.2 = 2 = 2 = 2 =
𝑛 𝑛 𝑛 𝑛
SRSWOR SRSWR

Mean of 𝑋 µ µ
2 2
Var of 𝑋 σ 𝑁−𝑛
( 𝑁−1 )
σ
𝑛 𝑛

Central Limit Theorem

2
σ
𝑋 ≈ 𝑁𝑜𝑟𝑚𝑎𝑙(µ, 𝑛
)

T-distribution, t-table 𝑋~𝑡(𝑣)

2 2 2
𝑥 distribution 𝑋~𝑥 (𝑣), 𝑥 table
2
i. 𝑃(𝑋 > 𝑥 α(𝑣)) = α
2
ii. 𝑃(𝑥 < 𝑋 α(𝑣)) = 1 − α

F distribution, 𝑋~𝐹(𝑣1, 𝑣2)

Sampling from a Normal Distribution