Page 1
Random Variables
Probability Distributions
Some material from ppt slides for Statistics for Managers Using Microsoft Excel, 5e © 2008 Pearson Prentice-Hall, Inc.
A probability distribution is a table of all possible outcomes for a variable and the associated probability.
Random variables assign numerical values to the outcomes • How to summarize a random variable?
– The mean of a probability distribution is called the expected value
– The standard deviation
Page 2
Random Variables
Discrete Random Variable
ContinuousRandom Variable
Probability Distributions
Some material from ppt slides for Statistics for Managers Using Microsoft Excel, 5e © 2008 Pearson Prentice-Hall, Inc.
A probability distribution is a table of all possible outcomes for a variable and the associated probability.
Random variables assign numerical values to the outcomes
Page 3
Random Variables
Discrete Random Variable
Probability Distributions
Some material from ppt slides for Statistics for Managers Using Microsoft Excel, 5e © 2008 Pearson Prentice-Hall, Inc.
A probability distribution is a table of all possible outcomes for a variable and the associated probability.
Random variables assign numerical values to the outcomes
Binomial distribution (n = 100, p = 0.3)
0.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.10
10 13 16 19 22 25 28 31 34 37 40 43 46 49 52
X
p(X
)
Counting the expected number of
successes by specifying the probability of
success
Binomial DistributionPoisson distribution (µ = 5)
0.00
0.020.04
0.06
0.080.10
0.12
0.14
0.160.18
0.20
0 2 4 6 8 10 12 14 16 18 20
X
p(X
)
Counting the expected number of arrivals by
specifying the approximate arrival
rate
Poisson Distribution
Page 4
Random Variables
Discrete Random Variable
ContinuousRandom Variable
Probability Distributions
Some material from ppt slides for Statistics for Managers Using Microsoft Excel, 5e © 2008 Pearson Prentice-Hall, Inc.
A probability distribution is a table of all possible outcomes for a variable and the associated probability.
Random variables assign numerical values to the outcomes
Uniform Distribution
Describes measurements that are equally likely
Exponential Distribution
Measurements of inter-arrival times
for the arrivals described by the Poisson process
Page 5
The Normal Distribution
• ‘Bell Shaped’ & Symmetric• Mean, Median and Mode are equal • Location is characterized by the mean, μ• Spread is characterized by the standard deviation, σ
The random variable has an infinite theoretical range: - to +
Mean = Median = Mode
p(X)
μ
σ
Some material from ppt slides for Statistics for Managers Using Microsoft Excel, 5e © 2008 Pearson Prentice-Hall, Inc.
2μ)(X
2
1
e2π
1p(X)
Page 6
The Normal Distribution
Some material from ppt slides for Statistics for Managers Using Microsoft Excel, 5e © 2008 Pearson Prentice-Hall, Inc.
Page 7
The Normal Distribution
Page 8
The next step would be to develop a probability distribution for your sample data (which would require that you calculate probabilities of all outcome categories)…
The convenience of named probability distributions is that determining probabilities can be accomplished with a formula or table (so you need to determine which probability distribution best describes your data)…
Page 9
Random Variables
Discrete Random Variable
Probability Distributions
Some material from ppt slides for Statistics for Managers Using Microsoft Excel, 5e © 2008 Pearson Prentice-Hall, Inc.
A probability distribution is a table of all possible outcomes for a variable and the associated probability.
Random variables assign numerical values to the outcomes
Binomial distribution (n = 100, p = 0.3)
0.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.10
10 13 16 19 22 25 28 31 34 37 40 43 46 49 52
X
p(X
)
Counting the expected number of
successes by specifying the probability of
success
Binomial Distribution
=BINOMDIST(k, n, p, 0 or 1)
Poisson distribution (µ = 5)
0.00
0.020.04
0.06
0.080.10
0.12
0.14
0.160.18
0.20
0 2 4 6 8 10 12 14 16 18 20
X
p(X
)
Counting the expected number of arrivals by
specifying the approximate arrival
rate
Poisson Distribution
=POISSON(k, l, 0 or 1)
Page 10
Random Variables
Discrete Random Variable
ContinuousRandom Variable
Probability Distributions
Some material from ppt slides for Statistics for Managers Using Microsoft Excel, 5e © 2008 Pearson Prentice-Hall, Inc.
A probability distribution is a table of all possible outcomes for a variable and the associated probability.
Random variables assign numerical values to the outcomes
Uniform Distribution
Describes measurements that are equally likely
=1/(b-a)
Exponential Distribution
Measurements of inter-arrival times
for the arrivals described by the Poisson process
=EXPONDIST(k, 1/l, 1)
Page 11
The Normal Distribution
Page 12
The Normal Distribution
m-s s-2s 2s-3s 3s
Page 13
The Normal Distribution
m-s s-2s 2s-3s 3s
Page 14
The Normal Distribution
m-s s-2s 2s-3s 3s
Page 15
The Normal Distribution
Some material from ppt slides for Statistics for Managers Using Microsoft Excel, 5e © 2008 Pearson Prentice-Hall, Inc.
To find P(m<a) • =NORMDIST(a, , m s, 1)
To find a: P(m<a) = p • =NORMINV(p, , ,1)m s
p(X)
μ
σ
2μ)(X
2
1
e2π
1p(X)
Page 16
The Standard Normal Distribution
• ‘Bell Shaped’ & Symmetric• Mean, Median and Mode are equal • The mean, μ, is always 0• The standard deviation, σ , is always 1
The std normal distr. tells how far from the mean the observation lies…
Some material from ppt slides for Statistics for Managers Using Microsoft Excel, 5e © 2008 Pearson Prentice-Hall, Inc.
Z
p(Z)
0
1
2
Z2
e2π
1p(Z)
To find P(z<a) • =NORMSDIST(
a)To find a: P(z<a) = p • =NORMSINV(
p)
Page 17
The Standard Normal Distribution
Some material from ppt slides for Statistics for Managers Using Microsoft Excel, 5e © 2008 Pearson Prentice-Hall, Inc.
Z
p(Z)
0
1
2
Z2
e2π
1p(Z)
To find P(z<a) • =NORMSDIST(
a)To find a: P(z<a) = p • =NORMSINV(
p)
…allows us to convert any set of measurements to a common scale
…standardizingTo standardize• =STANDARDIZE(a, , m
s)
• ‘Bell Shaped’ & Symmetric• Mean, Median and Mode are equal • The mean, μ, is always 0• The standard deviation, σ , is always 1
The std normal distr. tells how far from the mean the observation lies…
Page 18
The Standard Normal Distribution
0-1 1-2 2-3 3
Sure would be nice if my data were Normal…
Page 19
Sampling Distribution of the Sample Mean
• A sampling distribution of the sample mean is a collection of sample means (of a given sample size)
How should we describe the sampling distribution?
– The mean of a sampling distribution is the expected value, E(X)
– A measure of the variability in the mean from sample to sample is given by the Standard Error of the Mean
n
σσ
X
Page 20
Sampling Distribution of the Sample Mean
• A sampling distribution of the sample mean is a collection of sample means (of a given sample size)
• GOAL OF SAMPLING: for the mean of the sampling distribution to equal the true population mean
…an unbiased sample
E(X) =
Page 21
Central Limit Theorem
• If samples of a given size are selected from any population, the sampling distribution of the sample mean will be normally distr.
Population Distribution
μ
Some material from ppt slides for Statistics for Managers Using Microsoft Excel, 5e © 2008 Pearson Prentice-Hall, Inc.
Page 22
Central Limit Theorem
• If samples of a given size are selected from any population, the sampling distribution of the sample mean will be normally distr.
• As n gets larger and larger, this approximation gets better Sampling Distribution
(becomes normal as n increases)Larger sample sizeSmaller
sample size
xμ
Some material from ppt slides for Statistics for Managers Using Microsoft Excel, 5e © 2008 Pearson Prentice-Hall, Inc.
Page 23
Sampling from non-Normal Distributions
• For most distributions, n > 30 will give a sampling distribution that is nearly normal
• For fairly symmetric distributions, n > 15 will give a sampling distribution that is nearly normal
• For normal population distributions, the sampling distribution of the mean is always normally distributed
Some material from ppt slides for Statistics for Managers Using Microsoft Excel, 5e © 2008 Pearson Prentice-Hall, Inc.
Page 24
Sampling Distribution of the Sample Proportion
Proportions are also normally distributed…
• Expected value for the proportion: mp
= p
• Standard error for the proportion:n
)(1σp