distribution formulas Flashcards
(16 cards)
Conditions for binomial
fixed number of trials
independent of each other
constant probability of siuccess
two outcomes
Expectation and variance of binomial
E(X) = np
Var(X) = np(1-p)
conditions for poission
Random rate of occurcence
independent
uniform average rate of occurence
Expectation and variance of poission
Lambda for both
Formula for poission distribution
x !
conditions for the geometric
When there is a number of independent trials until a success is achieved
two outcomes
p(X=x) for geometric distribution
(1-p)x-1 x p
P(X <= x) For geometric
1 - (1 - p) x
expectation and variance
E(X) = 1 / P
Var(X) = 1 - p / p2
when to use uniform distribution
When there is an equally likely number of outcomes
Expectation and variance of a uniform distribution
E(X) = n + 1 / 2
Var(X) = ( n + 1)(n - 1) / 12
proof of uniform distribution expectation and variance
Assumptions for PMCC hypothesis test
data must be bivariate normal.
Random on random.
Scatter diagram should be elliptical without skew or bimodalness
data is linear
data is without outliers
Explain effect size
For large values of n even a small non zero PMCC value will lead to a rjection of the null hypothesis so use a general rule insted.
How to calculate degrees of freedom
Number of rows - 1, number of columns - 1
if just one row then just n - 1
if yiu had to estimate a variable for a poission or binomial goodness of fit test then you minus 1 again