Upload
others
View
3
Download
0
Embed Size (px)
Citation preview
2/1/2012
1
Chapter 3
Basic Concepts in Statistics
and Probability
3.6 Continuous Distributions
• Normal distribution
• Student t-distribution
• Exponential distribution
• Lognormal distribution
• Weibull distribution
• Extreme value distribution
• Gamma distribution
• Chi-square distribution
• Truncated normal distribution
• Bivariate and multivariate
normal distribution
• F distribution
• Beta distribution
• Uniform distribution
2/1/2012
2
The normal distribution (also called the Gaussian distribution) is by far the most commonly used distribution in statistics. This distribution provides a good model for many, although not all, continuous populations.
The normal distribution is continuous rather than discrete. The mean of a normal population may have any value, and the variance may have any positive value.
3
3.6.1 Normal Distributions
Probability Density Function, Mean,
and Variance of Normal Dist.
The probability density function of a normal
population with mean and variance 2 is
given by
If X ~ N(, 2), then the mean and variance
of X are given by
4
2 2
X
X
2 2( ) / 21( ) ,
2
xf x e x
2/1/2012
3
68-95-99.7% Rule
This figure represents a plot of the normal probability density
function with mean and standard deviation . Note that the
curve is symmetric about , so that is the median as well as
the mean. It is also the case for the normal population.
• About 68% of the population is in the interval .
• About 95% of the population is in the interval 2.
• About 99.7% of the population is in the interval 3.
5
Standard Units
• The proportion of a normal population that is
within a given number of standard deviations of
the mean is the same for any normal population.
• For this reason, when dealing with normal
populations, we often convert from the units in
which the population items were originally
measured to standard units.
• Standard units tell how many standard deviations
an observation is from the population mean.
6
2/1/2012
4
Standard Normal Distribution
In general, we convert to standard units by subtracting the mean and dividing by the standard deviation. Thus, if x is an item sampled from a normal population with mean and variance 2, the standard unit equivalent of x is the number z, where
z = (x - )/.
The number z is sometimes called the “z-score” of x. The z-score is an item sampled from a normal population with mean 0 and standard deviation of 1. This normal distribution is called the standard normal distribution.
7
Finding Areas Under the Normal
Curve • The proportion of a normal population that lies within a
given interval is equal to the area under the normal probability density above that interval. This would suggest integrating the normal pdf, but this integral does not have a closed form solution.
• So, the areas under the curve are approximated numerically and are available in Table B. This table provides area under the curve for the standard normal density. We can convert any normal into a standard normal so that we can compute areas under the curve.
• The table gives the area in the right-hand tail of the curve between and z. Other areas can be calculated by subtraction or by using the fact that the normal distribution is symmetrical and that the total area under the curve is 1.
8
2/1/2012
5
Normal Probabilities
Excel:
NORM.DIST(x, mean, standard_dev, cumulative)
NORM.INV(probability, mean, standard_dev)
NORM.S.DIST(z)
NORM.S.INV(probability)
Minitab:
Calc Probability Distributions Normal
9
Linear Functions of Normal
Random Variables
Let X ~ N(, 2) and let a ≠ 0 and b be constants.
Then aX + b ~ N(a + b, a22).
Let X1, X2, …, Xn be independent and normally distributed with means 1, 2,…, n and variances 1
2, 22,…, n
2. Let c1, c2,…, cn be constants, and c1 X1 + c2 X2 +…+ cnXn be a linear combination. Then
c1 X1 + c2 X2 +…+ cnXn
~ N(c11 + c2 2 +…+ cnn, c121
2 + c222
2 + … +cn2n
2)
10
2/1/2012
6
Distributions of Functions of Normal
Random Variables
Let X1, X2, …, Xn be independent and normally distributed
with mean and variance 2. Then
Let X and Y be independent, with X ~ N(X, X2) and
Y ~ N(Y, Y2). Then
11
2
~ , .σ
X N μn
2 2
2 2
~ ( , )
~ ( , )
X Y X Y
X Y X Y
X Y N μ μ σ σ
X Y N μ μ σ σ
3.6.2 t Distribution
• If X~N(, 2)
• If is Not known, but n30
• Let X1,…,Xn be a small (n < 30) random sample from
a normal population with mean . Then the quantity
has a Student’s t distribution with n -1 degrees of
freedom (denoted by tn-1). 12
n
XZ
/
)(
(3.7)
ns
XZ
/
)( (3.8)
ns
Xt
/
)( (3.9)
2/1/2012
7
More on Student’s t
• The probability density of the Student’s t
distribution is different for different degrees of
freedom.
• The t curves are more spread out than the
normal.
• Table C, called a t table, provides probabilities
associated with the Student’s t distribution.
13
t Distribution
The probability density function of t
distribution is given by
𝑓 𝑡 =1
(𝑛 − 1)𝜋
Γ(𝑛2)
Γ,𝑛 − 12
-(1 +
𝑡2
𝑛 − 1)−𝑛/2
−∞ < 𝑡 < ∞
14
2/1/2012
8
t Distribution
www.boost.org/.../graphs/students_t_pdf.png
Other uses of t Distribution
𝑡 =𝜃 − 𝜃
𝑠𝜃
Where is the parameter to be estimated,
𝜃 is the estimator based on a sample,
𝑠𝜃 is the sample estimator of 𝜎𝜃
16
2/1/2012
9
3.6.3 Exponential Distribution
• The exponential distribution is a continuous
distribution that is sometimes used to model the
time that elapses before an event occurs (life
testing and reliability).
• The probability density of the exponential
distribution involves a parameter, which is the
mean of the distribution, , whose value
determines the density function’s location and
shape.
17
Exponential R.V.:
pdf, cdf, mean and variance
The pdf of an exponential r.v. is
𝑓 𝑥 =1
𝜃𝑒−𝑥/𝜃 𝑥 > 0
The cdf of an exponential r.v. is 𝐹 𝑥 = 1 − 𝑒−𝑥/𝜃 𝑥 > 0
The mean of an exponential r.v. is
μx =
The variance of an exponential r.v. is
σx2 = 2.
18
(3.10)
2/1/2012
10
Exponential Probability Density
Function
19
=1/
Exponential Probabilities
Excel:
EXPONDIST(x, lambda, cumulative)
Minitab:
Calc Probability Distributions Exponential
20
2/1/2012
11
Example
A radioactive mass emits particles according to a
Poisson process at a mean rate of 15 particles
per minute. At some point, a clock is started.
1. What is the probability that more than 5 seconds
will elapse before the next emission?
2. What is the mean waiting time until the next
particle is emitted?
21
Lack of Memory Property
The exponential distribution has a property known
as the lack of memory property:
If T ~ Exp(1/), and t and s are positive
numbers, then
P(T > t + s | T > s) = P(T > t).
22
2/1/2012
12
Example
The lifetime of a transistor in a particular circuit has an exponential distribution with mean 1.25 years.
1. Find the probability that the circuit lasts longer than 2 years.
2. Assume the transistor is now three years old and is still functioning. Find the probability that it functions for more than two additional years.
3. Compare the probability computed in 1. and 2.
23
3.6.4 Lognormal Distribution
• For data that contain outliers, the normal distribution is generally not appropriate. The lognormal distribution, which is related to the normal distribution, is often a good choice for these data sets.
• If X ~ N(,2), then the random variable Y = eX has the lognormal distribution with parameters and 2.
• If Y has the lognormal distribution with parameters and 2, then the random variable X = lnY has the N(,2) distribution.
24
2/1/2012
13
Lognormal pdf, mean, and
variance
The pdf of a lognormal random variable with parameters
and 2 is
𝑓 𝑥 =1
2𝜋𝜎 1
𝑥 𝑒𝑥𝑝 −
1
2𝜎2(ln 𝑥 − 𝜇)2 𝑥 > 0
Mean:
Variance:
2/2
)( eXE
22 222)( eeYV
25
Lognormal Probability Density
Function
26
=0
=1
2/1/2012
14
Lognormal Probabilities
Excel:
LOGNORM.DIST(x, mean, standard_dev)
Minitab:
Calc Probability Distributions Lognormal
27
Example
When a pesticide comes into contact with the skin,
a certain percentage of it is absorbed. The
percentage that is absorbed during a given time
period is often modeled with a lognormal
distribution. Assume that for a given pesticide,
the amount that is absorbed (in percent) within
two hours is lognormally distributed with a mean
of 1.5 and standard deviation of 0.5. Find the
probability that more than 5% of the pesticide is
absorbed within two hours.
28
2/1/2012
15
3.6.5 Weibull Distribution
The Weibull distribution is a continuous random variable that is used in a variety of situations. A common application of the Weibull distribution is to model the lifetimes of components. The Weibull probability density function has two parameters, both positive constants, that determine the scale and shape. We denote these parameters (scale) and (shape).
If = 1, the Weibull distribution is the same as the exponential distribution.
The case where =1 is called the standard Weibull distribution
29
Weibull R.V.
The pdf of the Weibull distribution is
𝑓 𝑥 = 𝛼𝛽(𝛼𝑥)𝛽−1𝑒−(𝛼𝑥)𝛽 𝑥 > 0
The mean of the Weibull is
𝜇𝑥 =1
𝛼Γ(1 +
1
𝛽)
30
The variance of the Weibull is
𝜎𝑥2 =
1
𝛼2 Γ 1 +
2
𝛽− Γ(1 +
1
𝛽)
2
2/1/2012
16
Weibull Probability Density
Function
31
=1, =5
=0.5, =1 =0.2, =5
Weibull R.V.
The cdf of the Weibull distribution (with =1) is
𝐹 𝑥 = 1 − 𝑒−(𝑥)𝛽 𝑥 > 0
32
2/1/2012
17
Weibull Probabilities
Excel:
WEIBULL(x, alpha, beta, cumulative)
Minitab:
Calc Probability Distributions Weibull
33
3.6.6 Extreme Value Distribution
Extreme value distributions are often used in reliability work.
34
2/1/2012
18
Extreme Value Distribution
The pdf of the Extreme value distribution is
𝑓 𝑥 =1
𝑏exp ,
𝑥 − 𝑎
𝑏-exp *− exp
𝑥 − 𝑎
𝑏+ − ∞ < 𝑥 < ∞
with a the location parameter and b the scale parameter
The cdf of the Extreme value distribution is
𝐹 𝑥 = 1 − exp *−exp𝑥 − 𝑎
𝑏+
• The Weibull distribution is the limiting distribution of the
extreme value distribution
• The natural log of a Weibull random time is an extreme value
random observation. 35
Extreme Value Distribution
36 http://www.itl.nist.gov/div898/handbook/apr/section1/apr163.htm
2/1/2012
19
3.6.7 Gamma Distribution
The gamma distribution is frequently a probability model for waiting times; for instance, in life testing, the waiting time until death is a random variable that is frequently modeled with a gamma distribution.
Gamma function is defined by
𝑓 𝑥 =1
𝛽𝛼Γ(𝛼)𝑥𝛼−1𝑒−𝑥/𝛽 𝑥 > 0
37
(3.11)
Where >0 (shape) and >0 (scale)
Gamma Distribution
• Standard gamma distribution when =1
• Mean: Variance: 2
• If k is an integer, then the distribution represents an Erlang distribution.
• If = 1, the gamma distribution is the same as the exponential.
• The gamma function has the following properties:
1. If r is any integer, then (r) = (r – 1)!.
2. For any r, (r + 1) = r (r).
3. Γ1
2= π
4. Γ 𝑟 = 𝑥𝑟−1𝑒−𝑥𝑑𝑥∞
0
38
2/1/2012
20
Gamma Probability Density
Function
39
=1, =1
=3, =0.5
=5, =1
Gamma Probabilities
Excel:
GAMMA.DIST(x, alpha, beta, cumulative)
GAMMA.INV(probability, alpha, beta)
GAMMALN(x)
Minitab:
Calc Probability Distributions Gamma
40
2/1/2012
21
• If = r/2, and =2 where r is a positive integer, the distribution is called a chi-square distribution with r degrees of freedom.
• The chi-square distribution results when independent variables with standard normal distributions are squared and summed.
• The probability density function of the chi-square distribution is
𝑓 𝑥 =𝑒−𝑥2 𝑥
𝜈2−1
2𝜈2Γ(𝜈
2) 𝑥 ≥ 0
41
3.6.8 Chi-Squre Distribution
42
Chi-Squre Distribution
2/1/2012
22
3.6.9 Truncated Normal Distribution
• Left truncated
• Right truncated
• Doubly truncated
43
Truncated Normal Distribution
44
2/1/2012
23
Left Truncated Normal Distribution
• Left truncated normal distribution
𝑓 𝑥 =𝑘
𝜎 2𝜋𝑒−1/2,(𝑥−𝜇)/𝜎-
2 𝑥 ≥ 𝑎
Where
𝑘 = 1 − Φ(𝑎 − 𝜇
𝜎)−1
Where a is the truncation point, and is the
cumulative area under the normal curve.
45
Lift Truncated Normal Distribution
• Let c = (𝑎 − 𝜇 )/𝜎
• Expected value
𝐸 𝑥 = 𝜇 + 𝑘𝜎𝑓,𝑐-
Where
𝑘 = 1 −Φ(𝑐) −1
𝑓 𝑐 =1
2𝜋𝑒−
12𝑐2
• Variance 𝑉𝑎𝑟 𝑥 = 𝜎2*1 + 𝑐𝑓(𝑐),1 − Φ 𝑐 -−1 − ,𝑓(𝑐)-2,1 − Φ 𝑐 -−2+
46
(3.12)
2/1/2012
24
Right Truncated Normal Distribution
• Right truncated normal distribution
𝑓 𝑥 =𝑘′
𝜎 2𝜋𝑒−1/2,(𝑥−𝜇)/𝜎-
2 𝑥 ≥ 𝑎
Where
𝑘′ = Φ(𝑎′ − 𝜇
𝜎)
−1
Where a’ is the truncation point, and is the
cumulative area under the normal curve.
47
Right Truncated Normal Distribution
• Let c′ = (𝑎′ − 𝜇 )/𝜎
• Expected value
𝐸 𝑥 = 𝜇 − 𝑘′𝜎𝑓,𝑐′-
Where
𝑘 = Φ(𝑐′) −1
𝑓 𝑐′ =1
2𝜋𝑒−
12𝑐′2
• Variance 𝑉𝑎𝑟 𝑥 = 𝜎2*1 + 𝑐′𝑓(𝑐′),1 − Φ 𝑐′ -−1 − ,𝑓(𝑐′)-2,1 − Φ 𝑐′ -−2+
48
2/1/2012
25
3.6.10 Bivariate and Multivariate
Normal Distribution
• Normal distribution
• For 2 independent variables, x1~N(1,12),
x2~N(2,22)
𝑓 𝑥1, 𝑥2 =1
2𝜋𝜎1𝜎2exp,−(𝑥1−𝜇1)
2
2𝜎12 -−(𝑥2−𝜇2)
2
2𝜎22 -
49
2 2( ) / 21( ) ,
2
xf x e x
(3.13)
Bivariate Normal Distribution
50
2/1/2012
26
Bivariate Normal Distribution
• Bivariate Normal distribution
𝑓 𝑥1, 𝑥2 =1
2𝜋𝜎1𝜎2exp,−(𝑥1−𝜇1)
2
2𝜎12 -−(𝑥2−𝜇2)
2
2𝜎22 -
• Let 𝕩 =𝑥1𝑥2
, 𝕞 =𝜇1𝜇2
, 𝕊 =𝜎1
2 0
0 𝜎22 ,
𝑓 𝕩 =1
2𝜋|𝕊|1/2exp,−1
2(𝕩−𝕞)′𝕊−1(𝕩−𝕞)-
51
(3.13)
Multivariate Normal Distribution
• Let 𝕩 =
𝑥1⋮𝑥𝑝
, 𝕞 =
𝜇1⋮𝜇𝑝
, 𝕊 =𝜎1
2 ⋯ 0⋮ ⋱ ⋮0 ⋯ 𝜎𝑝
2
𝑓 𝕩 =1
(2𝜋)𝑝/2|𝕊|1/2exp,−1
2(𝕩−𝕞)′𝕊−1(𝕩−𝕞)-
52
(3.14)
2/1/2012
27
3.6.11 F Distribution
• Let W, and Y be independent 2 random variables with u and degrees of freedom, the ratio F = (W/u)/(Y/ ) has F distribution.
• Probability Density Function, with u, degrees of freedom,
• Mean
• Variance
x
xuu
xuu
xfu
uu
0
1)()2
()2
(
))(2
()(
2
122
2)2(
)(
forxE
4)4()2(
)2(2)(
2
2
for
u
uxV
F Distribution
• Table V in Appendix A.
u
uf
f,,
,,1
1
2/1/2012
28
3.6.12 Beta Distribution
Beta function is defined by
𝑓 𝑥 =𝑟+𝑠−1 !
𝑟−1 ! 𝑠−1 !𝑥𝑟−1(1 − 𝑥)𝑠−1 0 ≤ 𝑥 ≤ 1
Mean of Beta distribution: r/(r+s)
Variance of Beta distribution: rs(r+s)-2(r+s+1)-1
55
Where r and s are shape parameters of Beta distribution
3.6.12 Beta Distribution
56 http://astarmathsandphysics.com/university_maths_notes/probability_and_statistics/probability_
and_statistics_the_beta_distribution.html
2/1/2012
29
3.6.12 Beta Distribution
Beta distribution and F distribution
𝐵 𝛼; 𝑟, 𝑠 =2𝑟/2𝑠)𝐹(𝛼;2𝑟,2𝑠
1+(2𝑟/2𝑠)𝐹(𝛼;2𝑟,2𝑠)
57
Where B(; r, s) denotes the (1- ) percentile of a beta distribution
with parameters r and s.
3.6.13 Uniform Distributions
The uniform distribution has two parameters, a
and b, with a < b. If X is a random variable with
the continuous uniform distribution then it is
uniformly distributed on the interval (a, b). We
write X ~ U(a,b).
The pdf is
58
1,
( )
0, otherwise
a x bf x b a
2/1/2012
30
Uniform Distribution:
Mean and Variance
If X ~ U(a, b).
Then the mean is
and the variance is
59
2X
a bμ
22 ( )
.12
X
b aσ