Download Discrete Probability Distributions

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts
no text concepts found
Transcript
Discrete Probability
Distributions
Quantitative Methods II
Plan for Today
• Discrete and continuous random variables
• Probability distributions
• Histograms for discrete probability
distributions
• Examples
• The expected value, the variance, and the
standard deviation
• Lotteries (example: Lotto 6/49)
• Probability of winning, expected payoff,
payoff rate, expected winnings.
1
Random Variables
• We say that x is a random (quantitative)
variable if the value of x is determined by an
outcome of an experiment or an observation
and cannot be predicted precisely.
• A discrete random variable can have finitely
many or a countable number of values.
• A continuous random variable can take on
countless number of values on a given interval.
• Examples: 1. Number of passengers on a bus.
2. Weight of a watermelon.
Probability Distributions
• A probability distribution is an assignment of a
probability to each value or to each interval of
values of a random variable.
• For a discrete probability distribution (DPD),
each distinct value of a random variable has its
own probability.
• The sum of all probabilities equals 1.0000
• Let us consider an example:
Example: it is known that 25% of German adults
are smokers. A sample of 4 German adults is
selected at random.
2
Example: smokers
• We can have 0, 1, 2, 3, or 4 smokers in the
sample. The respective probabilities can be
computed (recall how!) and put in a table:
x
P(x)
0
1
2
3
4
0.3164 0.4219 0.2109 0.0469 0.0039
Check that σ 𝑃(𝑥) = 1 .
This table is an example of a discrete probability
distribution (DPD).
A histogram for a DPD
• The data in the table can be better visualized
using a histogram.
3
The expected value
• The mean of the probability distribution is
also called the expected value and is computed
using the following formula:
𝜇 = ෍ 𝑥 ∙ 𝑃(𝑥)
In the previous example,
𝜇 = 0 ∙ 0.3164 + 1 ∙ 0.4219 + 2 ∙ 0.2109 +
+ 3 ∙ 0.0469 + 4 ∙ 0.0039 = 1.0000
Why does it make sense to expect one smoker
out of four randomly selected Germans?
The variance and standard deviation
These quantities are computed as follows:
𝜎2 = σ 𝑥 − 𝜇
2
∙ 𝑃(𝑥)
and
𝜎 = 𝜎2
In the previous example,
𝜎 2 = 0 − 1.0 2 ∙ 0.3164 + 1 − 1.0 2 ∙ 0.4219
+ 2 − 1.0 2 ∙ 0.2109 + 3 − 1.0 2 ∙ 0.0469 +
+ 4 − 1.0 2 ∙ 0.0039 = 0.7500
and
𝜎 = 𝜎 2 = 0.8660
Why is it a reasonable value?
4
Example: marbles in a bag
There are 16 marbles in a bag: 6 yellow and 10
red. Four marbles are chosen at random and
the number of red marbles, x , is counted.
We have the following probability table
(explain how the numbers are obtained)
x
P(x)
0
1
2
3
4
0.0082 0.1099 0.3709 0.3956 0.1154
Again, check that σ 𝑃(𝑥) = 1.0000 (or close)
The histogram
5
Example (red and yellow balls): continued
• Compute the expected value, the variance,
and the standard deviation of this probability
distribution.
• μ = 2.5000
• Why does this number make perfect sense?
• 𝜎 2 = 0.7496
• 𝜎 = 0.8658
• Does this number seem reasonable?
Example: lottery 6/49
• For $3 a player chooses 6 numbers from 1 to 49.
• Six winning numbers plus 1 complimentary
number are drawn at random by the lottery
officials.
• There are seven winning categories and prizes:
W6 (Jackpot)
W5 + C
W5
W4
W3
W2 + C
W2
$ 10 000 000
(typical prizes listed)
$ 100 000
$ 2 000
$ 100
$ 10
$5
$ 3 (free play)
6
Computing probabilities for Lotto 6/49
For the category W2, where two winning
numbers are guessed correctly:
1
8
15
22
29
36
43
2
9
16
23
30
37
44
3
10
17
24
31
38
45
4
11
18
25
32
39
46
5
12
19
26
33
40
47
6
13
20
27
34
41
48
7
14
21
28
35
42
49
Computing probabilities for Lotto 6/49
The total number of possible choices is
49𝐶6 = 13 983 816
The number of choices for two winning numbers
are 6𝐶2 = 15
The number of choices for four losing numbers
are 42𝐶4 = 111 930
Thus, the probability of winning in this category:
𝑃 2𝑊 =
6𝐶2 ∙ 42𝐶4
= 0.1201
49𝐶6
7
Computing probabilities for Lotto 6/49
For the category W2+C, where 2 winning and the
complimentary number are guessed correctly:
1
8
15
22
29
36
43
2
9
16
23
30
37
44
3
10
17
24
31
38
45
4
11
18
25
32
39
46
5
12
19
26
33
40
47
6
13
20
27
34
41
48
7
14
21
28
35
42
49
Computing probabilities for Lotto 6/49
The number of choices for two winning numbers
are 6𝐶2 = 15
The number of choices for the complimentary
number: 1𝐶1 = 1
The number of choices for three losing numbers
are 42𝐶3 = 11 480
Thus, the probability of winning in this category:
𝑃 2𝑊 + 𝐶 =
6𝐶2 ∙ 1𝐶1 ∙ 42𝐶3
= 0.01231
49𝐶6
8
Probabilities for other categories:
6𝐶3 ∙ 43𝐶3
= 0.01765
49𝐶6
6𝐶4 ∙ 43𝐶2
𝑃 4𝑊 =
= 0.0009686
49𝐶6
6𝐶5 ∙ 42𝐶1
𝑃 5𝑊 =
= 0.00001802
49𝐶6
6𝐶5 ∙ 1𝐶1
𝑃 5𝑊 + 𝐶 =
= 0.000000429
49𝐶6
6𝐶6
𝑃 Jackpot =
= 0.000000071
49𝐶6
𝑃 3𝑊 =
Another way of formulating probabilities
One chance in 1 / P(x). Consider Lotto 6/49:
Category
2W
2W + C
3W
4W
5W
5W + C
6W
Chances
1 in 8.3
1 in 81.2
1 in 56.7
1 in 1033
1 in 55 492
1 in 2 330 636
1 in 13 983 816
9
Overall probability of winning
• It equals the sum of all winning probabilities.
• In the example of Lotto 4/49, we have
𝑃 winning = 0.1201 + 0.01231 + 0.01765 +
+ 0.0009686 + 0.00001802 + 0.000000429 +
+ 0.000000071 = 0.1510
or 1 in 6.6 (so the majority of all tickets contain
losing combinations)
The expected payoff
It is defined as the sum σ 𝑥 ∙ 𝑃(𝑥), where x is the
winning amount and P(x) is the corresponding
probability. In the example of Lotto 6/49:
෍ 𝑥 ∙ 𝑃(𝑥) = 3 ∙ 0.1201 + 5 ∙ 0.01231 +
+ 10 ∙ 0.01765 + 100 ∙ 0.0009686 +
+ 2000 ∙ 0.00001802 + 100 000 ∙ 0.000000429
+ 10 000 000 ∙ 0.000000071 =
= $1.48
10
The payoff rate
It is defined as follows:
The payoff rate =
Expected payoff
Price of ticket
=
1.48
3.00
= 0.49
It can also be expressed in terms of percentages
The payoff rate = 49% . This also indicates the
rate of return on your investment.
The expected winnings
They are defined as follows:
Expected winnings =
= Expected payoff – Price of ticket
In the example of Lotto 6/49,
Expected winnings = 1.48 – 3.00 = – $1.52
The negative sign indicates that the player is
actually expected to lose $1.52 with the
purchase of each ticket.
Is it profitable to play such a lottery?
11