Formulas Flashcards
Binomial Coefficient
m!/k!(m-k)!
Expectation of X
E(X) = Σxf(x)
Variance of X
V(X) = E(X^2) - μ^2
Standard Deviation of X
SD(X) = √σ
Linear Transformation (Formula)
E(Y) = …
V(Y) = …
SD(Y) = …
Y = a + bx E(Y) = a + bE(X) V(Y) = b^2 V(X) SD(Y) = I b I σ^2
Linear Transformation (Properties)
- V(a) = 0
2. E(a) = a
Standardization of a Random Variable (Formula)
Z = (X-μ) / σ
Standardization of a Random Variable (Properties)
E(Z) = 0 V(Z) = 1
Chebyshev’s Inequality in Probability (Formula)
P(μ - kσ < X < μ + kσ) > 1- 1/(k^2)
Bernouli Distribution (Formula)
E(X) = …
V(X) = …
E(X) = p V(X) = p(1-p)
Bernouli Distribution (Properties)
is used to model random variables with two outcomes
Y ~ Bin(n,p) (Formula)
(nCr) (p^k) ((1-p)^n-k)
Y ~ Bin(n,p) (Properties)
E(Y) = …
V(Y) = …
E(Y) = np V(Y) = np(1-p)
Proportion of Success (p̂) (Formula)
E(p̂) = …
V(p̂) = …
p̂ = #of successes/n E(p̂) = p V(p̂) = (p(1-p))/n
Hypergeometric Distribution (drawn with replacement)
Y ~ Bin(n,p) where p = #successes/n
E(Y) = …
V(Y) = …
E(Y) = np V(Y) = np(1-p)
Hypergeometric Distribution (drawn without replacement)
Y ~ Bin(n,#successes/N) only if n/N < 0.1
Uniform Distribution Y ~ U(α,β)
E(Y) = …
V(Y) = …
E(Y) = (α+ß)/2 V(Y) = (ß-α)^2/12
Normal Distribution (Formula)
Y ~ N(μ,σ^2)
E(Y) = …
V(Y) = …
E(Y) = μ V(Y) = σ^2
Normal Distribution (3 x Properties)
- pdf is symmetrical around μ
- f(x) max at μ
- probability = area under the graph
Approximation of μ (Formula)
P(x̄ - Z(a/2) σ < μ < x̄ + Z(a/2) σ) = 1 - a
Joint Probability Density Function
E(W) = …
COV(W) = …
E(W) = E(W(X,Y) = w(x,y) h(x,y) COV(W) = E((X-E(X))(Y-E(Y))
Expectation of a Sum
E(X) = …
V(X) = …
E(ΣX) = nE(X) V(ΣX) = nV(X)
Nominal Variables (Definition)
Values that cannot be ordered in a natural way
ex. (1) single, (2) married, (3) divorced, (4) widowed
Ordinal Variables (Definition)
Values can be ordered
ex. (1) Poor, (2) Average, (3) Rich
Qualitative Variables (Definition)
Values are categories and not numbers
Quantitative Variables (Definition)
Are numbers / Variables
Population Mean
(1/N) Σ X
Population Variance
(1/N) Σ(X^2) - μ^2
Sample Mean
(1/n) Σ X
Sample Variance
s^2 = (1/n-1) Σ (x-x̄)^2
Central Limit Theorem (Definition)
If n is larger than 30, then x̄ ≈ N(μ, σ^2/n)
Application only possible if np>5 and n(1-p)>5
Approximation of p
P(p̂ - Z(a/2) (√(p(1-p)/n) < p < p̂ + Z(a/2) (√(p(1-p)/n)) = 1 - a
5 Step Procedure for H0 testing
i) Formulate Testing Problem
ii) Test Statistic
iii) Rejection region
iv) calculate the VAL
v) draw conclusion of H0 and H1
T Value
Used when σ is unkown
ta/2;n-1
Regression Line (Formula)
Y = β0 + β1X + ε
Covariance of X and Y
cov(X,Y) = E[(x-μx)(y-μy)]
Expectation of X and Y
E(XY) = E(X)E(Y)
Variance of a Linear Combination
W = a + bx + cx
V(W) = b^2V(X) + c^2V(X) + 2bcCOV(XY)