Crib Sheet for Exam #1 Statistics 211
1 Chapter 1: Descriptive Statistics
Sample Average Population Average
x =
1
n
n
i=1
x
i
=
1
N
N
i=1
x
i
To caluclate the pth percentile x
[p]
:
1. Let x
(i)
refer to our data set in ascending order.
2. Let i
p
= np/100.
3. Find the rst index i such that i > i
p
.
4. The pth percentile is then:
x
[p]
=
x
(i1)
+x
(i)
2
if i 1 = i
p
x
(i)
otherwise
s
2
=
1
n 1
(x
i
x)
2
=
1
n 1
x
2
i
x
i
2
n
Chebychevs Rule: The proportion of observations that are within k standard
deviations (p
k
) of the mean is at least:
p
k
= 1
1
k
2
2 Chapter 2: Probability
Multiplication rule Permutation Combination
n
1
n
2
n
2
. . . n
k
P
k,n
=
n!
(nk)!
n
k
=
P
k,n
k!
=
n!
k!(nk)!
For any two events A and B:
P(A B) = P(A) +P(B) P(A B)
The conditional probability of A given that B occurred (P(B) > 0):
P(A|B) =
P(A B)
P(B)
Two events A and B are independent if
P(A|B) = P(A)
If A and B are independent then
P(A B) = P(A)P(B)
If A, B, C, D, . . . are mutually independent then
P(A B C D. . .) = P(A)P(B)P(C)P(D) . . .
1
3 Chapter 3: Discrete PDFs
E[X] = =
XS
x p(x)
E[h(X)] =
h(x)
=
XS
h(X) p(x)
E(aX +b) = aE(X) +b
V (X) =
2
= E[(x u)
2
]
=
XS
(x )
2
p(x)
= E[X
2
] E[X]
2
V (aX +b) = a
2
V (X) = a
2
2
3.1 Binomial Distribution
For X binomial(n, p)
n = xed number of trials
p = probability of succes (S)
x = number of successes (S)
P(X = x) =
n
x
p
x
(1 p)
nx
x = 0, 1, 2, . . . , n
= E[X] = np
2
= V [X] = E[(x )
2
] = np(1 p)
3.2 Multinomial Distribution
For X multinomial(n, p
1
, . . . , p
r
)
n = Number of trials.
r = Number of possible outcomes.
p
i
= P(Outcome i on any particular trial).
x
i
= Number of trials resulting in outcome i.
p(x
1
, x
2
, . . . , x
r
) =
n!
x
1
!x
2
! . . . x
r
!
p
x1
1
p
x2
2
. . . p
x
r
r
x
i
= 0, 1, 2, . . . x
1
+x
2
+. . . x
r
= r
2
3.3 Hypergeometric Distribution
For X hypergemetric(n, M, N)
n = sample size
M = number of S in the population
N = population size
x = number of S in the sample
P(X = x) =
M
x
NM
nx
N
n
where x : max(0, n N +M) x min(n, M).
E[X] = = n
M
N
= np where p =
M
N
V [X] =
2
= (
Nn
N1
)n
M
N
(1
M
N
)
= (
Nn
N1
)np(1 p)
3.4 Binomial Approximation to the Hypergeometric
If we sample with replacement of if n is small relative to N and M, we can
approximate the Hypergeometric distribution by using the binomial distribution
with p =
M
N
:
X hypergemetric(n, M, N) X binomial(n, p =
M
N
)
3.5 Hypergeometric Distribution for k Cells
N items are partitioned into k cells A
1
, A
2
, . . . , A
k
with a
1
, a
2
, . . . , a
k
elements
respectively. Then the probability distribution of the random variables X
1
, X
2
, . . . , X
k
representing the number of elements selected from A
1
, A
2
, . . . , A
k
in a random
sample of size n is:
P(X
1
= x
1
, X
2
= x
2
, . . . , X
k
= x
k
) =
a1
x
1
a2
x
2
. . .
a
k
x
k
N
n
where
k
i=1
x
i
= n
k
i=1
a
i
= N
For the case k = 2:
A
1
= S (success)
A
2
= F (failure)
a
1
= M (number of S in A
1
)
a
2
= N M (number of F in A
2
)
n = sample size
N = population size
3
3.6 Negative Binomial Probability Distribution
For X negative binomial(r, p)
r = number of S
p = probability of S
x = the number of failures preceding the rth success
P(X = x) =
x +r 1
r 1
p
r
(1 p)
x
x = 0, 1, 2, . . .
E[X] = =
r(1p)
p
V [X] =
2
=
r(1p)
p
2
If r = 1 we have a Geometric distribution.
P(X = x) = p(1 p)
x
x = 0, 1, 2, . . .
3.7 Poisson Distribution
For X poisson()
= the rate per unit time or rate per unit area.
x = the number of successes occurring during
a given time interval or in a specied region
P(X = x) =
e
x
x!
x = 0, 1, 2, . . . > 0
E[X] = =
V [X] =
2
=
3.8 Poisson Approximation to the Binomial Distribution
Let X be a binomial random variable with probability distribution X binomial(n, p).
When n and p 0 and = np remains xed at > 0, then
X binomial(n, p) X poisson( = np)
As a rule of thumb, this approximation can be safely applied if:
n 100 p .01 np 20
4
4 Chapter 4: Continuous PDFs
P(a X b) =
b
a
f(x)dx
F(x) = P(X x) =
f(y)dy
P(a X b) = F(b) F(a)
F
(X) = f(x)
For 0 p 1 the (100p)th percentile of a continuous distribution you must
solve p = F(x) for x where x is the (100p)th percentile.
E[h(x)] =
h(x) f(x)dx
E[X] = =
x f(x)dx
E[(X )
2
] =
2
=
(x u)
2
f(x)dx
Remember: E[(X )
2
] = E[X
2
] E[X]
2
=
2
4.1 The Uniform Distribution
The family of uniform distributions has the following PDF:
f(x, A, B) =
1
BA
A x B
0 otherwise
4.1.1 The Exponential Distribution
The family of exponential distributions has the following PDF:
f(x, ) =
e
x
x 0, > 0
0 otherwise
5
