Statistics Summary

PERMUTATIONS AND COMBINATIONS

The number of permutations of r objects, taken from a set of n distinct objects without replacement is given by n! n Pr = (n-r)!

PROBABILITY
PROBABILITY For two events A and B, P(A B)=P(A)+P(B)-P(A B) P(A)=P(A B)+P(A B') MUTUAL EXCLUSIVITY

The number of permutation of r objects, taken from a set of n distinct objects with replacement is given by nr The number of permutations of n distinct objects in a circle is given by (n-1)! The number of possible combinations of r objects, taken from a set of n distinct objects without replacement is given by n! n Cr = (n-r)!r!

Mutually exclusive events cannot occur at the same time. For two mutually exclusive events, E1 and E2, P(E1 E2 )=P(E1 )+P(E2 )
CONDITIONAL PROBABILITY

P(A|B)= INDEPENDENCE

P(A B) P(B)

Independent events are events the occurrences of which do not influence the probability of the occurrence of the other event. For independent events, P(A|B)=P(A) or P(B|A)=P(B) P(A B)=P(A)P(B)

1 of 3

Statistics Summary

RANDOM VARIABLES
For any random variable X, The expectation, , is given by E(X)= xP(X=x) and E(aXbY) = aE(X) bE(Y) The variance is given by
Var(X)=E[(X-)2 ]= (x-) P(X=x)
2

CONTINUOUS RANDOM VARIABLES

NORMAL DISTRIBUTI ON For a random variable X modelled by a normal distribution with mean and standard deviation 2 X~N(, ) STANDARD NORMAL VARIABLE Letting X~N(,2), the standard normal variable Z is defined as x- Z= ~N(0,1) and
2

and Var(aXbY)=a Var(X) + b Var(Y) The standard deviation, , is given by = Var(X)

P(Xx) = P(Z

x- )

DISCRETE RANDOM VARIABLES

BINOMIAL DISTRIBUTIO N For a random variable X modelled by a binomial distribution with n trials and probability of success, p X~B(n,p) Its probability distribution is given by n x n-x P(X=x)= Cxp (1-p) Its mean and variance are given by E(X) = np Var(X) = np(1-p) POISSON DISTRIBUTION For a random variable X modelled by a Poisson distribution with parameter X~Po() Its probability distribution is given by e- x P(X=x)= x! Its mean and variance are given by E(X) = Var(X) = Note also that for two Poisson random variables X~Po(1) and Y~Po(2), X+Y~Po(1+ 2)

APPROXIMATIONS
Approximations marked are to be continuity corrected

BINOMIAL TO POISSON For X~B(n,p) If n is large (n > 50) and p is small (p < 0.1) such that np < 5, then X~Po(np) BINOMIAL TO NORMAL For X~B(n,p) In is large such that np > 5 and n(1-p) > 5, then X~N(np,np(1-p)) POISSO N TO NORMAL For X~Po() If > 10, then X~N(, ) CONTINUITY CORRECTION These are the ranges, for given probability distribution functions, to consider when approximating discrete random variables to continuous random variables. P(Xa) P(X<a) P(x=a) P(X>a) P(Xa) a-1 a a+1

2 of 3

Statistics Summary

SAMPLING
SAMPLE MEAN The sample mean from a normal population 2 of sample size n with mean and variance is given by 2 X~N(, ) n CENTRAL LIMIT THEOREM The central limit theorem states that, for a non-normal population with sample size n, 2 X~N(, ) approximately, if n is large (50). n UNBIASED ESTIMATOR OF SAMPLE MEAN For any sample size n taken from a population with an unknown mean , the unbiased estimator of is given by x = (x-a) + a x= n n where a is a constant
UNBIASED ESTIMATOR OF SAMPLE VARIANCE For any sample size n taken from a population with an unknown mean 2, the unbiased estimator of 2 is given by 2 x 1 1 2 2 ( ) x s = = (x-x)2 n-1 n n-1 2 (x-a) ) 1 2 ( = (x-a) - n n-1 where a is a constant

HYPOTHESIS TESTING
CONDUCTING A HYPO THESIS TEST Step 1: Step 2: Step 3: Step 4: Step 5: State the null and alternative hypotheses H0 and H1 State the significance level, Determine the test statistic to use and its distribution Calculate the p-value for the test statistic Indicate whether or not to reject H0 based on the evidence from the sample H0 is rejected if p-value < H0 is not rejected if p-value >

TEST STATISTICS Normal Population 2 known 2 unknown Sample size is large n50 Test Statistic Sample size is small Test statistic Non-normal Population 2 known 2 unknown by the CLT by the CLT 2 s2 X~N(, ) X~N(, ) n n Z-test Z-test

X~N(,

2 ) n

X~N(,

s2 ) n

Z-test 2 X~N(, ) n Z-test

Z-test T~t(n-1) t-test

3 of 3