4.1 probability distributions
TRANSCRIPT
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 1/32
4.1 Probability
Distributions
Statistics
Mrs. Spitz
Fall 2010
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 2/32
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 3/32
Random Variables
• The outcome of a probability experiment is often a
count or a measure. When this occurs, the
outcome is called a random variable.
• A random variable, x, represents a numerical
value assigned to an outcome of a probability
experiment.• There are two types: discrete and continuous
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 4/32
Discrete v. continuous
• A random variable is discrete if it has a
finite or countable number of possibleoutcomes that can be listed.
• A random variable is continuous if it has an
infinite number of possible outcomes
represented by an interval on the number
line.
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 5/32
So . . .
• Suppose you conduct an experiment of thenumber of calls a salesperson makes in oneday. The possible values of the random variable
are 0, 1, 2, 3, 4, and so on. Because the set ofpossible outcomes {0, 1, 2, 3 . . . } can be listed,x is a discrete random variable. You canrepresent its values as points on a number line.
0 1 2 3 4 5 6 7 8 x can only have wholenumber values 0, 1, 2, 3 . . .
. . .
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 6/32
So . . .
• A different way to conduct the study would be tomeasure the time (in hours) a salespersonspends making calls in one day. Because thetime spent making sales calls can be any
number from 0 to 24 (including fractions anddecimals), x is a continuous random variable.You can represent its values with an interval ona number line, but you cannot list all the possiblevalues.
0 3 6 9 12 15 18 21 24 x can have any valuebetween 0 and 24
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 7/32
Ex. 1: Discrete variables andcontinuous variables.
• Decide whether the random variable, x, isdiscrete or continuous. Explain your reasoning.
1. x represents the number of stocks in the Dow
Jones Industrial Average that have share priceincreases on a given day.
The number of stocks whose share valueincreases can be counted {0, 1, 2, 3 . . . }. So xis a discrete random variable.
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 8/32
Ex. 1: Discrete variables andcontinuous variables.
• Decide whether the random variable, x, isdiscrete or continuous. Explain yourreasoning.
2. x represents the volume of bottled waterin a 32-ounce container.
The amount of water in the container can beany volume between 0 and 32, so x is acontinuous random variable.
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 9/32
Note:
• It is important that you can distinguish between discrete and continuous variables
because different statistical techniques areused to analyze each. The remainder of thischapter focuses on discrete random
variables and their probability distributions.You will study continuous distributionslater.
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 10/32
Discrete Probability Distributions
• Each value of a discrete random variable can beassigned a probability. By listing each value of therandom variable with its corresponding probability, youare forming a probability distribution.
1. The probability of each value of the discrete randomvariable is between 0 and 1 inclusive. That is,
0 P(x) 1
2. The sum of all the probabilities is 1. That is,
P(x) = 1
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 11/32
Graphing
• Because probabilities represent relative
frequencies, a discrete probability
distribution can be graphed with a relative
frequency histogram.
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 12/32
Guidelines of Constructing
a Discrete Probability
DistributionLet x be a discrete random variable with possible
outcomes x1, x2, . . . xn.
1. Make a frequency distribution for the possibleoutcomes.
2. Find the sum of the frequencies.
3. Find the probability of each possible outcome by
dividing the frequency by the sum of thefrequencies.
4. Check that each probability is between 0 and 1and that the sum is 1.
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 13/32
Ex. 2: Constructing a DiscreteProbability Distribution
• An industrial psychologist has administered apersonality inventory test for passive-aggressivetraits to 150 employees. Individuals were rated
on a score from 1 to 5 where 1 was extremelypassive and 5 extremely aggressive. A score of3 indicated neither trait. The results are shownon the next slide. Construct a probability
distribution for the random variable, x. Thengraph the distribution.
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 14/32
Solution
Divide the frequency ofeach score by the totalnumber of scores (150) tofind the probability for
each value of the randomvariable.
P(1) = 24/150 = 0.16
P(2) = 33/150 = 0.22
P(3) = 42/150 = 0.28
P(4) = 30/150 = 0.2
P(5) = 21/150 = 0.14
Score (x) Frequency (f)
1 24
2 33
3 42
4 30
5 21
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 15/32
• The discreteprobabilitydistribution is shown
in the followingtable. Note thateach probability isbetween 0 and 1and the sum of the
probabilities is 1.
• The relativefrequencydistribution is also
shown at the right.The area of eachbar represents theprobability of aparticular outcome.
0
0.05
0.1
0.15
0.2
0.25
0.3
1 2 3 4 5
RF
x 1 2 3 4 5
P(x) 0.16 0.22 0.28 0.2 0.14
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 16/32
Ex. 3: Verifying ProbabilityDistributions
• Verify that thedistribution is aprobability
distribution.
Days of Rain Probability
0 0.216
1 0.432
2 0.288
3 0.064
Solution: If the distributionis a probability
distribution, the (1) eachof probability is between0 and 1, inclusive and (2)the sum of theprobabilities equals 1.
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 17/32
Ex. 3: Verifying ProbabilityDistributions
1. Each probability isbetween 0 and 1. Days of Rain Probability
0 0.216
1 0.432
2 0.288
3 0.064
2. P(x) = 0.216 + 0.432 +
0.288 + 0.064 = 1
Because both conditions are
met, the distribution is aprobability distribution.
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 18/32
Ex. 4: Probability Distributions
• Decide whether each distribution is a probabilitydistribution.
x 5 6 7 8
P(x) 0.28 0.21 0.43 0.15
Each probability is
between 0 and 1.
However, the sum of
the probabilities is1.07, which is greaterthan 1. So, it is NOTa probabilitydistribution.
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 19/32
Ex. 4: Probability Distributions
• Decide whether each distribution is a probabilitydistribution.
x 1 2 3 4
P(x) ½ ¼ 5/4 -1
However, P(3) andP(4) are not between0 and 1. So, it is NOTa probabilitydistribution.
The sum of theprobabilities is equalto 1.
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 20/32
Mean, Variance and Standard
Deviation• You can measure the central tendency of a
probability distribution with its mean, and measurethe variability with its variance and standard
deviation.
The mean of a discrete random variable is given by:
= xP(x).
Each value of x is multiplied by its corresponding probability and the products are added.
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 21/32
Note:
• The mean of the random variable represents
the “theoretical average” of a probabilityexperiment and sometimes is not a possible
outcome. If the experiment were performed
thousands of times, the mean of all the
outcomes would be close to the mean of therandom variable.
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 22/32
Ex. 5: Finding the Mean of a ProbabilityDistribution
The probabilitydistribution for thepersonality inventory
test for passive-aggressive traitsdiscussed in Ex. 2 isgiven at the right.
Find the mean score.What can youconclude?
x P(x)
1 0.16
2 0.22
3 0.28
4 0.20
5 0.14
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 23/32
Organize your tables carefully.
• Use the table to organizeyour work as shown atthe left. From the table,you can see that the
mean is 2.94. A score of3 represents an individualwho is neither extremelypassive nor aggressive,
but is slightly closer topassive.
x P(x) xP(x)
1 0.16 1(0.16) = 0.16
2 0.22 2(0.22) = 0.44
3 0.28 3(0.28) = 0.84
4 0.20 4(0.20) = 0.80
5 0.14 5(0.14) = 0.70
P(x) = 1 xP(x) = 2.94MEAN
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 24/32
Note:
• While the mean of the random variable of a
probability distribution describes a typical
outcome, it gives no information about how
the outcomes vary. To study the variation
of the outcomes, you can use the variance
and standard deviation of the randomvariable of a probability distribution.
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 25/32
Standard Deviation of a Discrete RandomVariable
• The variance of a discrete randomvariable is:
2 = (x - )2P(x)
The standard deviat ion is:
= √2
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 26/32
Ex. 6: Finding the variance and StandardDeviation
• The probabilitydistribution for thepersonality inventorytest for passive-aggressive traitsdiscussed in Ex. 2 isgiven at the right.Find the variance and
standard deviation ofthe probabilitydistribution.
x P(x)
1 0.16
2 0.22
3 0.28
4 0.20
5 0.14
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 27/32
x P(x) x - (x -)2 P(x)(x - )2
1 0.16 1 – 2.94 =
-1.94
(-1.94)2 =
3.764
(0.16)(3.764) = 0.602
2 0.22 2 – 2.94 =
- 0.94
(-0.94)2 =
0.884
(0.22)(0.884) = 0.194
3 0.28 3 – 2.94 =
0.06
(0.06)2 =
0.004
(0.28)(0.004) = 0.001
4 0.20 4 – 2.94 =
1.06
(1.06)2 =
1.124
(0.20)(1.124) = 0.225
5 0.14 5 – 2.94 =
2.06
(2.06)2 =
4.244
(0.14)(4.244) = 0.594
P(x) = 1 P(x)(x - )2 = 1.616 =
VARIANCE
You know from ex. 5 that the mean of the distribution is = 2.94. Use a tablelike the one below to organize your work!
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 28/32
• So, the variance of 2 = 1.616 and thestandard deviation is
= √1.62 ≈1.27
Lots of steps, so organize your workcarefully. Most of you got the problemscorrect, but some missed the details.
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 29/32
Expected Value
• The expected value of a discrete randomvariable is equal to the mean of therandom variable.
Expected value = E (x) = = xP (x)
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 30/32
Ex. 7: Finding an Expected Value
• At a raffle, 1500 tickets are sold at $2 eachfor four prizes of $500, $250, $150 and$75. You buy one ticket. What is the
expected value of your gain?
• Note: Expected value plays a role indecision theory. Although probability can
never be negative, the expected value canbe negative.
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 31/32
Gain, x $498 $248 $148 $73 - $2
Probability, P(x)
1/1500 1/1500 1/1500 1/1500 1496/1500
SOLUTION: To find the gain for each prize, subtract theprice of the ticket from the prize. For instance, your gain for
the $500 prize is $500 - $2 = $498. Then write a probabilitydistribution for the possible gains (or outcomes).
8/11/2019 4.1 Probability Distributions
http://slidepdf.com/reader/full/41-probability-distributions 32/32
Then, using the probability distribution, you can find theexpected value. E(x) = xP(x)
35.1$
1500
1496)2(
1500
173
1500
1148
1500
1248
1500
1498
)(E(x)
x xP
Because the expected value is negative, you can expectto lose an average of $1.35 for each ticket you buy.