1530 6 probability distributions
TRANSCRIPT
-
7/27/2019 1530 6 Probability Distributions
1/18
'
&
$
%
Lesson 6: Probability Distributions
1 Distributions 2
2 Binomial Random Variables 6
3 Normal Random Variables 8
http://sites.google.com/site/mattjones1204/ 1
-
7/27/2019 1530 6 Probability Distributions
2/18
'
&
$
%
1 Distributions
A random variable is a quantitative variable whose value depends on chance.
Well use them to model the behavior of populations.
A discrete random variable is a random variable whose possible values form
a finite or countably infinite set of numbers.
A continuous random variable is a random variable whose possible values
form a continuum or interval of numbers.
Use capital letters for random variables and lower case letters for their
possible values.
The distribution of a random variable is TWO things:
1. a list of the possible values it can take on;
2. a list of the probabilities of those possible values.
http://sites.google.com/site/mattjones1204/ 2
-
7/27/2019 1530 6 Probability Distributions
3/18
'
&
$
%
Example 1 Let X = # of siblings for a person chosen at random.
# of Siblings x 0 1 2 3 4
P(X = x) 0.2 0.425 0.275 0.075 0.025
X is the random variable representing the number of siblings.
x represents a possible value for the # of siblings and can be 0, 1, 2, 3, or 4.
P(X = x) is the probability that the number of children equals x.
For any discrete random variable X,x
P(X = x) = 1
because the events {X = x} are disjoint and the probability of getting someoutcome in the sample space is 1.
http://sites.google.com/site/mattjones1204/ 3
-
7/27/2019 1530 6 Probability Distributions
4/18
'
&
$
%
Mean of a Random Variable
The mean or expected value of X is
= E[X] =x
xP(X = x)
The mean is a weighted average of the possible values of x, where the
weights are the probabilities.
If the distribution of X describes a population, the sample means of large
samples tend to be close to the mean .
Example 2 Find the expected number of siblings for a person chosen at
random.
# of Siblings x 0 1 2 3 4
P(X = x) 0.2 0.425 0.275 0.075 0.025
http://sites.google.com/site/mattjones1204/ 4
-
7/27/2019 1530 6 Probability Distributions
5/18
'
&
$
%
Standard Deviation of a Discrete Random Variable
If there are N equally likely possible outcomes for a random variable X (so
each occurs with probability 1/N), the standard deviation of X is given by
=
x(x )2
N=
x
(x )2 1/N
If the outcomes are not equally likely, this generalizes to
=
x
(x )2P(X = x)
Example 3 Find the standard deviation of the number of siblings.
# of Siblings x 0 1 2 3 4
P(X = x) 0.2 0.425 0.275 0.075 0.025
http://sites.google.com/site/mattjones1204/ 5
-
7/27/2019 1530 6 Probability Distributions
6/18
'
&
$
%
2 Binomial Random Variables
X is a binomial random variable with parameters n and p if
P(X = x) =
n
x
px(1 p)nx
where x is any number in
{0, 1, 2, . . . , n
}. X denotes the number of successes
that occur in n independent trials where each trial has success probability p.
Example 4 There are 49 senators, each of whom shows for the senate
meeting with probability 0.95. What is the probability that no one shows?
What is the probability that everyone shows? What is the probability that
exactly 46 show? What is the probability that at least 47 show?
http://sites.google.com/site/mattjones1204/ 6
-
7/27/2019 1530 6 Probability Distributions
7/18
'
&
$
%
Mean and Standard Deviation of a Binomial Random Variable
If X is a binomial random variable with parameters n and p, its mean is
given by
= np
and its standard deviation is given by
=np(1 p)Example 5 Of the 49 senators, each showing up with probability 0.95
independently of the others, what is the mean number who show? What is
the standard deviation of the number who show?
http://sites.google.com/site/mattjones1204/ 7
-
7/27/2019 1530 6 Probability Distributions
8/18
'
&
$
%
3 Normal Random Variables
A random variable has the normal distribution with mean and standard
deviation if its density function is
f(x) =1
2e(x)2
22
where is the mean and is the standard deviation. The graph of thedensity function looks like this:
http://sites.google.com/site/mattjones1204/ 8
-
7/27/2019 1530 6 Probability Distributions
9/18
'
&
$
%
More about Normal Random Variables
Normal random variables arise in the study of natural and industrial systems.
Many statistical tests are based on normal distributions.
If X has a normal distribution with mean and standard deviation , then
P(a X b) is the area of the region under the normal curve between thepoints a and b above the x-axis:
If X is distributed normal with mean and standard deviation , use the
calculator to find probabilities by doing
P(a
X
b) = normalcdf(a,b,,)
http://sites.google.com/site/mattjones1204/ 9
-
7/27/2019 1530 6 Probability Distributions
10/18
'
&
$
%
Example 6The amount of coffee X your statistics professor drinks eachday has a normal distribution with mean 55 oz and standard deviation 6 oz.
Shade the appropriate regions beneath the density function and find the
following:
P(49
X
55) =
P(55 X 61) =P(X 55) =P(X > 65) =
P(X < 45 or X > 65) =
P(|X 55| 12) =
http://sites.google.com/site/mattjones1204/ 10
-
7/27/2019 1530 6 Probability Distributions
11/18
'
&
$
%
Empirical Rule
The empirical rule states if X is distributed normal with mean and
standard deviation , then
P(|X | ) 68.27%P(|X | 2) 95.45%P(|X | 3) 99.73%
Another way to think about this: for a normally distributed population with
mean and standard deviation , approximately
68.27% of all observations are within one
standard deviation of the mean,
95.45% of all observations are within two
standard deviations of the mean,
99.73% of all observations are within three
standard deviations of the mean.
http://sites.google.com/site/mattjones1204/ 11
-
7/27/2019 1530 6 Probability Distributions
12/18
'
&
$
%
Example 7 The heights of men in the United States are distributed
approximately normal with mean 69 inches and standard deviation 2.8
inches. Then the empirical rule tells us
Example 8 Ball bearings have masses distributed normal with mean 1.2
grams and standard deviation 0.03 grams. Then the empirical rule tells us
http://sites.google.com/site/mattjones1204/ 12
-
7/27/2019 1530 6 Probability Distributions
13/18
'
&
$
%
Standard Normal Random Variables
A standard normal random variable Z has a normal distribution with
mean 0 and standard deviation 1.
Setting = 0 and = 1 in the density function on Slide 2 gives the density
function for a standard normal random variable:
f(z) =1
2ez
2/2
The graph of the density function looks like this:
If Z is distributed normal with mean 0 and standard deviation 1, use your
calculator to find probabilities as
P(a Z b) = normalcdf(a,b,0,1) or simply normalcdf(a,b)
http://sites.google.com/site/mattjones1204/ 13
-
7/27/2019 1530 6 Probability Distributions
14/18
'
&
$
%
Relationship Between Standard Normals and Other Normals
If X is distributed normal with mean and standard deviation , thenX
has a normal distribution with mean 0 and standard deviation 1. That is, for
any numbers a and b,
P(a X b) = Pa
Z b
This observation allows us to use normal tables to obtain probabilities for
any normal random variable.
http://sites.google.com/site/mattjones1204/ 14
-
7/27/2019 1530 6 Probability Distributions
15/18
'
&
$
%
Example 9 Suppose X normal(5, 2). Calculate the following with yourcalculator or using normal tables:
P(X > 6) =
P(2.5
X
6.5) =
P(|X 5| < 2.2) =
P(X < 4.5) =
P(X < 3.1 or X > 5.5) =
http://sites.google.com/site/mattjones1204/ 15
-
7/27/2019 1530 6 Probability Distributions
16/18
'
&
$
%
z Notation
z is the number on the real line such that the probability of a standard
normal random variable taking on a greater value is :
P(Z > z) =
We find these values using normal tables in reverse, or using a calculator:
z = invNorm() = invNorm(1 - )
http://sites.google.com/site/mattjones1204/ 16
-
7/27/2019 1530 6 Probability Distributions
17/18
'
&
$
%
Example 10 Find the following with (1) your calculator, (2) using a normal
table. Also mark locations of the z-scores beneath density functions:
z0.025 =
z0.05 =
z0.5 =
z0.975 =
z0.62 =
z0.12 =
http://sites.google.com/site/mattjones1204/ 17
-
7/27/2019 1530 6 Probability Distributions
18/18
'
&
$
%
Example 11 Evidence suggests IQs are distributed approximatelynormal(100, 16).
1. Find the 90th percentile.
2. Find the 84th percentile.
3. Find the IQ such that only 5% of the population has higher IQs.
http://sites.google.com/site/mattjones1204/ 18