Content Notes Flashcards
Sample Space Ω=
{a,b,c,…}, events are defined e.g. 1,2,3 or S=success, F=failure
Discrete r.v:
A specific value in a set e.g. {1,2,3}
Continuous r.v:
A continuous value in a set e.g. (0,1)
fx(x)=
P(X=x)
Fx(x)=
P(X≤x)
P(X≤b)=
P(X≤a) + P(a<X≤b)
Fx(b)=
Fx(a) + P(a<X≤b)
P(a<X≤b)=
Fx(b) - Fx(a)
P(X≤10)=
Fx(10)
P(X<10)=
Fx(9)
2 conditions of a Bernoulli trial:
A succession of random independent experiments that must have,
1. Only 2 outcomes, success or failure
2. Constant probability, p
Bernoulli Trial pmf:
Y={0, if failure (1-p); 1, if success (p)
Probability function of a Bernoulli trial:
fY(y)=P(Y=y)={p, if y=1; 1-p, if y=0
Binomial Distribution fx(x)=
P(X=x)=(n x)p^x(1-p)^n-x
P(A|B)=
P(AnB)/P(B)
Bayes Theorem:
P(B|A)=(P(A|B)P(B))/P(A)
Partition Rule for a Bernoulli trial:
P(S)=P(S|T)P(T) + P(S|Tc)P(Tc)
Likelihood rule:
L(p;x)=P(X=x)=(n x)p^x(1-p)^n-x for 0≤p≤1
Maximum Likelihood Estimate:
dL(p;x)/dp =0, p=p̂, p̂=x/n
Maximum Likelihood Estimator:
To find the maximum value of p we take L(p) and derive it to get dL/dP, the max value occurs when p=0.
Var(p̂)=
Var(x/n)=1/n^2 Var(X) = 1//n^2 np(1-p) = p(1-p)/n
Var(aX)=
a^2(Var(X))
Var(X)=
np(1-p) = E(X^2)-(E(X))^2
Var(aX+b)=
a^2(Var(X))
E(X)=
np
sd(X)=
sqrt(Var(X))=sigma(x)
Standard error of p̂=
sqrt((p̂(1-p̂))/n)
What is the 95% confidence interval?
The true value is within 2 standard errors 95% of the time.
Distribution mean:
E(X)=Sum of xP(X=x) = Sum of xfx(x) = EX = ux
Binomial Distribution mean:
E(X) = Sum of x(n x)p^x(1-p)^n-x = np
What is a Monte Carlo simulation?
A statistical technique that predicts possible outcomes of an event.
What are the 2 basic ideas of the Monte Carlo simulation?
- Repeat an experiment r times
- The relative frequency of the value observed is the approximate probability of that value, e.g. If we see X=10 when N=50, then our p=0.2.
Finite population of size N, variance=
1/N Sum of (xi-u)^2
Population mean=
(1/N) Sum of xi
Estimator for Sn^2:
1/n Sum of (xi-xbar)^2
Estimator for Sn-1^2
1/n-1 Sum of (xi-xbar)^2
geometric distribution
P(X=x)= (1-p)^x*p