Discrete probability distributions Flashcards
Conditions for binomial distribution
Fixed number of trials
fixed chance of success
two outcomes only
independent success rate of trials
conditions of poisson distribution
mean rate of occurrence in a given time period
random event
events are independent of each other
Binomial notation
X ~ B(n, p)
where n is the number of trials and p is the chance of success
Binomial formula
P (X = r) = nCr x p^r x q^n-r
E(X) for binomial distribution
E(X) = np
Var(X) for binomial distribution
Var(X) = npq
or np - np^2
poisson formula
P(X = r) = e^-λ x λ^r/r!
where λ = the mean rate of occurrence
poisson notation
X ~ Po(λ)
λ i.e mean rate of occurrence
poisson skew
positively skewed
has less of an effect as λ increases
what is a unique feature of poisson distribution
E(X) = Var(X) = λ
good for questions which ask what type of distribution is appropriate
If X ~ Po(x)
and Y ~ Po(y)
Then what is X + Y
X + Y ~ Po(x + y)
Geometric distribution criteria
1 outcome, fixed and independent chance
expectation and variance of x
E = 1/p
Var = (1-p)/ p^2
Geometric distribution p values
P(X = x) - (1-p)^x-1 x p
P(X <= x) = 1-(1-p)^x
P(X > x) = (1-p)^x
Uniform distribution criteria
All outcomes have an even chance
only works if evenly distributed from a to b
Expectation and variance of x = 1….n
E = n+1)/2
Var = 1/12(n+1)(n-1)
Working out E and Var for x = a….b
E = a+b)/2
Var (and for E) - scale up the original formula 1…n then multiply / add var and e accordingly