Introduction to Probability

Distributions
Random Variable
A random variable x takes on a defined set of
values with different probabilities.
For example, if you roll a die, the outcome is random
(not fixed) and there are 6 possible outcomes, each of
which occur with probability one-sixth.
For example, if you poll people about their voting
preferences, the percentage of the sample that responds
Yes on Proposition 100 is a also a random variable (the
percentage will be slightly differently every time you
poll).
Roughly, probability is how frequently we
expect different outcomes to occur if we
repeat the experiment over and over
(frequentist view)
Random variables can be
discrete or continuous
Discrete random variables have a countable
number of outcomes
Examples: Dead/alive, treatment/placebo, dice,
counts, etc.
Continuous random variables have an
infinite continuum of possible values.
Examples: blood pressure, weight, the speed of a
car, the real numbers from 1 to 6.
Probability functions
A probability function maps the possible
values of x against their respective
probabilities of occurrence, p(x)
p(x) is a number from 0 to 1.0.
The area under a probability function is
always 1.
Discrete example: roll of a die
x
p(x)
1/6
1 4 5 6 2 3

=
x all
1 P(x)
Probability mass function (pmf)
x p(x)
1 p(x=1)=1/6
2 p(x=2)=1/6
3 p(x=3)=1/6
4 p(x=4)=1/6
5 p(x=5)=1/6
6 p(x=6)=1/6
1.0
Cumulative distribution function
(CDF)
x
P(x)
1/6
1 4 5 6 2 3
1/3
1/2
2/3
5/6
1.0
Cumulative distribution
function
x P(xA)
1 P(x1)=1/6
2 P(x2)=2/6
3 P(x3)=3/6
4 P(x4)=4/6
5 P(x5)=5/6
6 P(x6)=6/6
Review Question 1
If you toss a die, whats the probability that you
roll a 3 or less?
a. 1/6
b. 1/3
c. 1/2
d. 5/6
e. 1.0
Review Question 2
Two dice are rolled and the sum of the face
values is six? What is the probability that at
least one of the dice came up a 3?
a. 1/5
b. 2/3
c. 1/2
d. 5/6
e. 1.0
Review Question 2
Two dice are rolled and the sum of the face
values is six. What is the probability that at least
one of the dice came up a 3?
a. 1/5
b. 2/3
c. 1/2
d. 5/6
e. 1.0
How can you get a 6 on two
dice? 1-5, 5-1, 2-4, 4-2, 3-3
One of these five has a 3.
1/ 5
Continuous case
The probability function that accompanies
a continuous random variable is a
continuous mathematical function that
integrates to 1.
For example, recall the negative exponential
function (in probability, this is called an
exponential distribution): x
e x f

= ) (
1 1 0
0
0
= + = =
+

x x
e e
This function integrates to 1:
Continuous case: probability
density function (pdf)
x
p(x)=e
-x
1
The probability that x is any exact particular value (such as 1.9976) is 0;
we can only assign probabilities to possible ranges of x.
For example, the probability of x falling within 1 to 2:
23 . 368 . 135 . 2) x P(1
1 2
2
1
2
1
= + = = = =

e e e e
x x
x
p(x)=e
-x
1
1 2
Clinical example: Survival times
after lung transplant may
roughly follow an exponential
function.
Then, the probability that a
patient will die in the second
year after surgery (between
years 1 and 2) is 23%.
Example 2: Uniform
distribution
The uniform distribution: all values are equally likely.
f(x)= 1 , for 1 x 0
x
p(x)
1
1
We can see its a probability distribution because it integrates
to 1 (the area under the curve is 1):
1 0 1 1
1
0
1
0
= = =

x
Example: Uniform distribution
Whats the probability that x is between 0 and ?
P( x 0)=
Clinical Research Example:
When randomizing patients in
an RCT, we often use a random
number generator on the
computer. These programs work
by randomly generating a
number between 0 and 1 (with
equal probability of every
number in between). Then a
subject who gets X<.5 is control
and a subject who gets X>.5 is
treatment.
x
p(x)
1
1

0
Expected Value and Variance
All probability distributions are
characterized by an expected value
(mean) and a variance (standard
deviation squared).
Expected value of a random variable
Expected value is just the average or mean () of
random variable x.
Its sometimes called a weighted average
because more frequent values of X are weighted
more highly in the average.
Its also how we expect X to behave on-average
over the long run (frequentist view again).
Expected value, formally

=
x all
) ( ) p(x x X E
i i
Discrete case:
Continuous case:
dx ) p(x x X E
i i

=
x all
) (
Symbol Interlude
E(X) =
these symbols are used interchangeably
Example: expected value
Recall the following probability distribution of
ER arrivals:
x 10 11 12 13 14
P(x) .4 .2 .2 .1 .1

=
= + + + + =
5
1
3 . 11 ) 1 (. 14 ) 1 (. 13 ) 2 (. 12 ) 2 (. 11 ) 4 (. 10 ) (
i
i
x p x
Sample Mean is a special case of
Expected Value
Sample mean, for a sample of n subjects: =
)
1
(
1
1
n
x
n
x
X
n
i
i
n
i
i

=
=
= =
The probability (frequency) of each
person in the sample is 1/ n.
Expected Value
Expected value is an extremely useful
concept for good decision-making!
Example: the lottery
The Lottery (also known as a tax on people
who are bad at math)
A certain lottery works by picking 6 numbers
from 1 to 49. It costs $1.00 to play the
lottery, and if you win, you win $2 million
after taxes.
If you play the lottery once, what are your
expected winnings or losses?
Lottery
8 -
49
6
10 x 7.2
816 , 983 , 13
1
! 6 ! 43
! 49
1 1
= = =

x$ p(x)
-1 .999999928
+ 2 million 7.2 x 10
--8
Calculate the probability of winning in 1 try:
The probability function (note, sums to 1.0):
49 choose 6
Out of 49
numbers, this is
the number of
distinct
combinations of 6.
Expected Value
x$ p(x)
-1 .999999928
+ 2 million 7.2 x 10
--8
The probability function
Expected Value
E(X) = P(win)*$2,000,000 + P(lose)*-$1.00
= 2.0 x 10
6
* 7.2 x 10
-8
+ .999999928 (-1) = .144 - .999999928 = -$.86
Negative expected value is never good!
You shouldnt play if you expect to lose money!
Expected Value
If you play the lottery every week for 10 years, what are your
expected winnings or losses?
520 x (-.86) = -$447.20
Gambling (or how casinos can afford to give so
many free drinks)
A roulette wheel has the numbers 1 through 36, as well as 0 and 00.
If you bet $1 that an odd number comes up, you win or lose $1
according to whether or not that event occurs. If random variable X
denotes your net gain, X=1 with probability 18/38 and X= -1 with
probability 20/38.
E(X) = 1(18/38) 1 (20/38) = -$.053
On average, the casino wins (and the player loses) 5 cents per game.
The casino rakes in even more if the stakes are higher:
E(X) = 10(18/38) 10 (20/38) = -$.53
If the cost is $10 per game, the casino wins an average of 53 cents per
game. If 10,000 games are played in a night, thats a cool $5300.
Expected value isnt
everything though
Take the hit new show Deal or No Deal
Everyone know the rules?
Lets say you are down to two cases left. $1
and $400,000. The banker offers you
$200,000.
So, Deal or No Deal?
Deal or No Deal
This could really be represented as a
probability distribution and a non-
random variable:
x$ p(x)
+1 .50
+$400,000 .50
x$ p(x)
+$200,000 1.0
Expected value doesnt help
x$ p(x)
+1 .50
+$400,000 .50
x$ p(x)
+$200,000 1.0
000 , 200 ) 50 (. 000 , 400 ) 50 (. 1 ) (
x all
= + + = = =

) p(x x X E
i i

000 , 200 ) ( = = X E
How to decide?
Variance!
If you take the deal, the variance/ standard
deviation is 0.
If you dont take the deal, what is average
deviation from the mean?
Whats your gut guess?
Variance/standard deviation

2
=Var(x) =E(x-)
2
The expected (or average) squared
distance (or deviation) from the mean

= = =
x all
2 2 2
) ( ] ) [( ) ( ) p(x x x E x Var
i i

Variance, continuous

=
x all
2
) ( ) ( ) p(x x X Var
i i

Discrete case:
Continuous case?:
dx ) p(x x X Var
i i

=
x all
2
) ( ) (
Symbol Interlude
Var(X)=
2
SD(X) =
these symbols are used interchangeably
Similarity to empirical variance
The variance of a sample: s
2
=
)
1
1
( ) (
1
) (
2
1
2
1

=
=
n
x x
n
x x
N
i
i
N
i
i
Division by n-1 reflects the fact that we have lost a
degree of freedom (piece of information) because
we had to estimate the sample mean before we could
estimate the sample variance.
Variance

=
x all
2 2
) ( ) p(x x
i i

000 , 200 000 , 200
000 , 200 ) 5 (. ) 000 , 200 000 , 400 ( ) 5 (. ) 000 , 200 1 (
) (
2
2 2 2
x all
2 2
= =
= + =
= =

) p(x x
i i
Now you examine your personal risk tolerance
Important discrete probability
distribution: The binomial
Binomial Probability
Distribution
A fixed number of observations (trials), n
e.g., 15 tosses of a coin; 20 patients; 1000 people
surveyed
A binary outcome
e.g., head or tail in each toss of a coin; disease or no
disease
Generally called success and failure
Probability of success is p, probability of failure is 1 p
Constant probability for each observation
e.g., Probability of getting a tail is the same each time
we toss the coin
Binomial distribution
Take the example of 5 coin tosses.
Whats the probability that you flip
exactly 3 heads in 5 coin tosses?
Binomial distribution
Solution:
One way to get exactly 3 heads: HHHTT
Whats the probability of this exact arrangement?
P(heads)xP(heads) xP(heads)xP(tails)xP(tails)
=(1/2)
3
x (1/2)
2
Another way to get exactly 3 heads: THHHT
Probability of this exact outcome = (1/2)
1
x (1/2)
3
x (1/2)
1
= (1/2)
3
x (1/2)
2
Binomial distribution
In fact, (1/2)
3
x (1/2)
2
is the probability of each
unique outcome that has exactly 3 heads and 2
tails.
So, the overall probability of 3 heads and 2 tails
is:
(1/2)
3
x (1/2)
2
+ (1/2)
3
x (1/2)
2
+ (1/2)
3
x (1/2)
2
+ .. for as many unique arrangements as
there arebut how many are there??
Outcome Probability
THHHT (1/2)
3
x (1/2)
2
HHHTT (1/2)
3
x (1/2)
2
TTHHH (1/2)
3
x (1/2)
2
HTTHH (1/2)
3
x (1/2)
2
HHTTH (1/2)
3
x (1/2)
2
HTHHT (1/2)
3
x (1/2)
2
THTHH (1/2)
3
x (1/2)
2
HTHTH (1/2)
3
x (1/2)
2
HHTHT (1/2)
3
x (1/2)
2
THHTH (1/2)
3
x (1/2)
2
10 arrangements x (1/2)
3
x (1/2)
2
The probability
of each unique
outcome (note:
they are all
equal)
ways to
arrange 3
heads in
5 trials

5
3
5
C
3
= 5!/3!2! = 10
Factorial review: n! = n(n-1)(n-2)
P( 3 heads and 2 tails) = x P(heads)
3
x P(tails)
2
=
10 x ()
5=
31.25%

5
3
x
p(x)
0 3 4 5 1
2
Binomial distribution
function:
X= the number of heads tossed in 5 coin
tosses
number of heads
p(x)
number of heads
Binomial distribution,
generally
X n X
n
X
p p

) 1 (
1-p = probability
of failure
p =
probability of
success
X = #
successes
out of n
trials
n = number of trials
Note the general pattern emerging if you have only two possible
outcomes (call them 1/0 or yes/no or success/failure) in n independent
trials, then the probability of exactly X successes=
Binomial distribution: example
If I toss a coin 20 times, whats the
probability of getting exactly 10 heads?
176 . ) 5 (. ) 5 (.
10 10
20
10
=

Binomial distribution: example

=
= = =

+ = = =

+ = =

x
x x x
x x x
x
**All probability distributions are
characterized by an expected value and a
variance:
If X follows a binomial distribution with
parameters n and p: X ~ Bin (n, p)
Then:
E(X) = np
Var (X) = np(1-p)
SD (X)=
) 1 ( p np
Note: the variance will
always lie between
0*N-.25 *N
p(1-p) reaches
maximum at p=.5
P(1-p)=.25
Proportions
The binomial distribution forms the basis of
statistics for proportions.
A proportion is just a binomial count divided
by n.
For example, if we sample 200 cases and find 60
smokers, X=60 but the observed proportion=.30.
Statistics for proportions are similar to
binomial counts, but differ by a factor of n.
Stats for proportions
For binomial:
) 1 (
) 1 (
2
p np
p np
np
x
x
x
=
=
=

For proportion:
n
p p
n
p p
n
p np
p
p
p
p
) 1 (
) 1 ( ) 1 (

2
2

=
=

P-hat stands for sample