Binomial Probability Distribution

Report
Binomial Probability Distribution
Here we study a special discrete PD
(PD will stand for Probability
Distribution) known as the Binomial
PD
1
Row 0 --------------------1
Row 1------------------1
1
Row 2 ----------------1 2
1
Row 3 -------------1
3
3
1
Row 4 ---------1
4
6
4
1
What I have here is something called Pascal’s Triangle. Note
a) Each row starts and ends with a 1,
b) Any number that is not a 1 in the table is found by adding
values in the two blocks directly above it. For example, in row 2
the 2 = 1 + 1. Another example, in row 4 the 6 = 3 + 3.
c) A person could keep adding rows and the 5th row would have
the numbers 1, 5, 10, 10, 5, 1.
d) In any row the sum of the values = 2 raised to the power of the
row number. For example, in row 3 2^3 = 8. (the ^ symbol 2
means raise to the power).
I believe Pascal’s Triangle can help us understand the binomial
distribution. Now I want to start with a classic example of a
binomial distribution where a coin is flipped 4 times.
Since I picked flipping a coin 4 times I would look at row 4 of the
triangle. The numbers in the row have meaning for us. When you
flip a coin once it can be either heads or tails (on the side you see
after it lands). Let’s say heads is our focus of what we can see. Then
when we flip a coin 4 times the number of heads could be 0, 1, 2, 3,
and 4.
The number of heads in 4 flips:
0
1
2
3
4
Numbers in row 4:
1
4
6
4
1
So, when there are 0 heads in the 4 flips the number 1 below the 0
means there is only 1 way for this to happen. That would be TTTT,
or all 4 flips come up tails.
3
On the previous slide when I say the number of heads on 4 flips is
1, the number 4 below this means that there are 4 ways to have 1
head: TTTH, TTHT, THTT, HTTT.
Similarly, 2 heads in 4 flips can happen 6 ways: TTHH, THTH,
THHT, HTHT, HTTH, HHTT.
We could do this for 3 heads and 4 heads. (Will you do it?)
Now, check out the formula n!/k!(n-k)!. n is the number of flips
and k is the number of heads, since we said heads was our focal
point. The ! symbol means factorial. When you have a number
followed by a factorial sign you multiple the number by the next
lowest number all the way down to one. For example, 4! = 4
times 3 times 2 times 1 = 4(3)(2)(1) = 24.
4
0 heads in 4 flips happens 4!/0!(4!) = 1 way (since 0! = 1).
1 head in 4 flips happens 4!/1!(4-1)! = 4(3)(2)(1)/1(3)(2)(1) = 4
ways.
2 heads in 4 flips happens 4!/2!(4-2)! = 12/2 = 6 ways.
3 heads in 4 flips happens 4!/3!(4-3)! = 4 ways.
4 heads in 4 flips happens 4!/4!(4-4)! = 1 way.
The next thing we want to do is construct the binomial probability
distribution of the number of heads in 4 flips.
5
Binomial Distribution
When you flip a coin you can get a head or a tail. You could observe
heads or tails on several flips of a coin. Say you flip the coin n times
(n is a general number of times and when we have a specific problem
we usually have a specific value for n).
On each flip we might call heads the “event of interest” and after n
flips we might be interested in how many of the n flips gave the event
of interest. The possibilities for the number of events of interest take
on the discrete values 0, 1, 2 all the way through n. Thus the binomial
distribution is really just the distribution of a variable with discrete
values from 0 to n. But, certain conditions must hold. I show those
soon.
Now, on a coin flip heads has probability .5, but the event of interest on
any one trial in the more general binomial process does not have to be
6
.5.
Properties of a Binomial process.
1) The sample consists of a fixed numbers of observations, n.
2) Each observation is classified into one of two mutually exclusive
and collectively exhaustive categories.
3) The probability of an observation being classified as the event of
interest or a success, denoted by p, is constant (does not change) from
observation to observation. q = (1 - p) is the probability of an
observation being classified as NOT being the event of interest (or a
failure) and does not change from observation to observation.
4) The observations are independent.
Recall before we saw that the probability of the intersection of
independent events is equal to the product of the probabilities of each
event.
7
Let’s move to an example to put this information into context.
Say I wear a hat and every now and then I ask a person, “do you
like my hat?” Say in recent history that 10 percent of the people
said yes! So, thinking the future will be like the past (we will go
with this idea here), the likelihood any person would say yes (a
success) is 0.10.
So, a person saying yes is the event of interest and the probability
is .1 that this will happen to with any one person.
If a sample of 4 people are asked about my hat then the binomial
variable the number of folks saying yes can take on the values 0,
1, 2, 3, and 4.
There are 16 possible ways (2^4) the 4 orders can come in. Let’s
see these on the next screen.
8
Let’s call s the event of interest (a success) and f not an event of
interest (a failure).
4 orders
number of events of interest
ssss
sssf
ssfs
ssff
sfss
sfsf
sffs
sfff
fsss
fssf
fsfs
fsff
ffss
ffsf
fffs
ffff
4
3
3
2
3
2
2
1
3
2
2
1
2
1
1
0
On the next slide I have a tree
diagram to help you think about
all the possible outcomes. On the
far left I have the event of interest
s and not event of interest f in the
first round. The second round
shows a new s and f for each from
the previous round. Then you just
follow each branch to get the 16
different orders shown here.
9
S
S
S
F
S
F
S
F
S
S
F
F
S
F
F
S
S
S
F
F
F
S
F
S
F
S
F
F
S
F
10
The random variable again is x = the number of people saying
yes. So we can have 0, 1, 2, 3, or 4. You can see that 1 of the 16
possible outcomes has exactly 4 saying yes. Does this mean the
probability of exactly 4 saying yes is 1/16 or .0625? Maybe not.
Here is why. We have to figure in the probability that a given
person will say yes.
All 4 orders saying yes is really the intersection of the first saying
yes and the second saying yes and the third saying yes and the
fourth saying yes. The probability of the intersection of
independent events is just the multiplication of the probability of
each.
P(4) = .1(.1)(.1)(.1) = .0001 (By the way, if we flip a coin 4
times and the prob of a head on any 1 flip is .5, then the prob of 4
heads on 4 flips is .5(.5)(.5)(.5) = .0625 – BUT, here a person
saying yes only has prob = .1).
11
Now, the probability that only three say yes is tricky. I mean
tricky only until you see what is next.
You can see from the list 4 of the 16 had only 3 say yes. Each
one of the 4 has probability .1(.1)(.1)(.9) = .0009. So with 4
possible ways of getting this result we have P(3) = .0036
6 of the 16 outcomes have only two saying yes. Each has
probability .1(.1)(.9)(.9) = .0081. So with this occurring six
times P(2) = .0486.
4 of the 16 outcomes have only 1 saying yes. Each has
probability .1(.9)(.9)(.9) = .0729. So with this occurring 4
times P(1) = .2916.
1 of the 16 outcomes has none saying yes. P(0) = ,9(.9)(.9)(.9)
= .6561
12
Remember we have n = 4 people here and X = the number of
people saying yes to liking my hat. X could be 0, 1, 2, 3, or 4.
In general, the probability, written P(k), of a given k is found by
the formula
n!
pk(1-p)(n-k)
remember something raised to 0 power = 1
k!(n-k)!
Now in our example n = 4, p = .1 and 1-p = .9
When k = 0, the probability P(0) = 4!
.10(.9)4-0 = .6561
0!4!
Note when we found P(0) we had 0! and this equals 1. Plus we
had something raised to the 0 power. This always equals 1.
13
As you can tell these calculations are quite tedious. The good
news is our book has a table that can give us the probabilities
we so desire. Table C in the book has some binomial tables. Note
down the left side of the table you see examples of n from 2
through 20. Also on the left you see k as the number of items of
interest. For our example we had n = 4, or 4 people asked about
my hat and k represents the number saying yes. So we see the
probability that 0 of the 4 saying yes is in the 0 row (of the n = 4
section). Since in one trial our event of interest = .1 we have to
look in that column.
On the next screen I show you the binomial probability distribution
with n = 4 and p = .1. I also add the cumulative distribution.
On the next screen I have column P(X) and in the table in the book
k represents the values of X.
14
X
P(X)
0
.6561
1
.2916
2
.0486
3
.0036
4
.0001
P(X ≤ a value under the X column)
cumulative prob
The cumulative prob column is
.6561
telling us the prob of X in a
given row or any X less than in
.9477
the row.
.9963
P(X≤2), for example, is the
.9999
probability of 2 or fewer saying
yes and equals .9963
1.000
Now, let’s ask a few more questions.
What is the probability 1 or fewer say yes to liking my hat? The
cumulative prob tells us the answer is .9477.
What is the probability that more than 2 will say yes? More
than 2 is the complement of 2 or fewer, so P(more than 2) = 1 –
15
P(X≤2) = 1 - .9963 = .0037.
Microsoft Excel and the Binomial PD
On the next slide there is a spreadsheet in Excel. I use a different
generic example for you to see how this is similar to the table E.6.
Note cell c1 has the value of n = 3 and cell c2 has the value of p =
.3. Cells A4:A7 have the values of x.
Cells B4:B7 have Excel formulas typed in. If we put the mouse in
cell B4 and typed “=BINOMDIST(A4, $C$1, $CD$2, FALSE).”
The A4 will mean 0. The $C$1 will mean 3. The $C$2 will mean
.3 and the FALSE means we want the f(x). When you type this in
hit the enter key. To get the rest of the f(x) values put the mouse
back into cell B5 and click once. Then move the mouse to the
bottom right corner of the cell, click and drag down to the last cell.
In the BINOMDIST function A4 changes to A5 and so on as you
drag down. Excel wants to change cell values when you drag
functions. The $ signs in the $D$1 mean when you drag you will
16
not leave that cell. If you want a cum prob put TRUE, not FALSE.
17
18
The expected value for the binomial PD is
E(x) = np (a simplification for the binomial case from what we
saw previously for a discrete random variable), and the variance
is Var(x) = σ2 = np(1-p) (also a simplification).
The standard deviation is just the square root of the variance.
Consider a binomial experiment with n = 10 and p = 0.1. You can
double click inside the spreadsheet on the next screen and copy
the Excel file if you want.
a. P(0) is found in the f(x) column as .34867
b. P(2) = .1937
c. P(x≤2) is found in the Cum Prob column as .9298
d. P(x≥1) = 1 – P(x≤0) = 1 - .3487 = .6513
19
Number of Trials (n)
Probability of Success (p)
x
0
1
2
3
4
5
6
7
8
9
10
f(x)
0.34867844
0.387420489
0.193710245
0.057395628
0.011160261
0.001488035
0.000137781
8.748E-06
3.645E-07
9E-09
1E-10
10
0.1
Cum Prob
0.34867844
0.736098929
0.929809174
0.987204802
0.998365063
0.999853097
0.999990878
0.999999626
0.999999991
1
1
E(x) =
Var(x) =
St dev. =
1
0.9
0.948683
Note the E notation here. 9E-09 means we have the number 9 but have to
move the decimal 9 places to the left because we have E-09. The number is
.000000009. An E+ would require a movement of the decimal to the right.20
Note on the previous slide I have an Excel spreadsheet. At the
top I typed the label and numbers
Number of Trials (n) 10
Probability of Success (p)
0.1
in separate cells. The numbers are used in the formulas. You
should do this as well when you do a problem because it “dresses
up” the output and makes it easier to remember what the heck is
going on.
Also note that in my notes when you see a table you can double
click on it and see the Excel spreadsheet.
21
Example flipping two coins
If you flip two coins (or one coin twice) the possible outcomes
are HH, HT, TH, TT. So, n = 2. Let’s say the event of interest
is heads H. We could have X = 0, 1, or 2. Also say p = .5
From table c we see
X
P(X)
0
.25
1
.5
2
.25
What is the probability of at least 1 head on the two flips?
This would be P(1) + P(2) = .5 + .25 = .75
22
Special scenario
The binomial table c in the book only has probabilities up to 0.50.
But, in the real world probabilities can be up to 1.00.
Say you have a special coin that when flipped has a 0.60 probability
of being heads. Let’s flip 3 times. Thus there can be 0, 1, 2, or 3
heads in the 3 flips. Another way to state this is there can be 3, 2, 1,
or 0 tails in the 3 flips. The probability of a tail on any flip is .4 in
this case.
So, on the next slide I show you how to work with probabilities
greater then .5 in the binomial situation.
23
X = the number of heads and the probability of a head on any 1
flip is 0.6.
Y = the number of tails and the probability of a tail on any 1 flip
is 0.4.
X
P(X)
Y P(Y)
0
.064
3 .064
1
.288
2 .288
2
.432
1 .432
3
.216
0 .216
So, if the probability of a head on any one flip is 0.6 then use the
0.4 probability and note that 0 heads is like 3 tails (when you do
only 3 flips). 1 heads is like 2 tails, and so on.
As a last point I use the coin flip example a lot, but when there
are situations where we focus on many occurrences, but each
occurrence has only 2 outcomes, then we have the binomial dist.24
The Normal Distribution Approximation to the Binomial
Distribution
In our book we have binomial tables that go up to n = 20. But
often we have more than n = 20 as our interest. Some folks
figured out in these cases we can use the normal distribution to
give us a roughly equal answer.
Remember the normal distribution is characterized by a mean and
a standard deviation and then we use the Z calculation of
(value – mean) / standard deviation and round the result to 2
decimal places and then go to the z table to get probabilities.
In the binomial situation remember we have n trials and the
probability of a success on any 1 trail is p. In the normal
approximation the mean = n times p and the standard deviation is
the square root of the product of the 3 terms n, p, and (1 – p).
25
This normal approximation to the binomial can be used when 2
conditions hold:
n times p ≥ 10 and n times (1 – p) ≥ 10.
Example from book
5.69 Internet postings. Suppose (as is roughly true) that 20% of
all Internet users have posted photos online. A sample survey
interviews an SRS of 1555 Internet users.
(a) What is the actual distribution of the number X in the sample
who have posted photos online?
--the distribution is actually binomial with n = 1555 and p = .2
(b) What is the probability that 300 or fewer of the people in the
sample have posted photos online? (Use software or a suitable
approximation.) To use the normal approximation we have a mean
value = 1555 times .2 = 311 and a standard deviation equal to the
square root of (1555 times .2 times .8) = 15.77 see next slide for
26
more.
To get the probability of 300 or fewer we find the Z =
(300 – 311)/15.77 = -0.70 and the area to the left that is in the
table is what we want – that is .2420
So, when n is large (more than 20 for us) and the 2 conditions
noted are met, we can use the normal approximation to the
binomial distribution.
27

similar documents