78
STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Embed Size (px)

Citation preview

Page 1: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

STATISTICSUnivariate Distributions

Professor Ke-Sheng ChengDepartment of Bioenvironmental Systems Engineering

National Taiwan University

Page 2: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Probability density functions of discrete random variables

• Discrete uniform distribution • Bernoulli distribution• Binomial distribution• Negative binomial distribution• Geometric distribution• Hypergeometric distribution• Poisson distribution

04/10/23 2Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 3: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Discrete uniform distribution

N ranges over the possible integers.

)(1

0

,,2,11

);( ,,2,1 xINotherwise

NxNNxf NX

2/)1(][ NXE

N

j

jtX N

etm

NXVar

1

2

1)(

12/)1(][

04/10/23 3Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 4: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Bernoulli distribution

1-p is often denoted by q.

)()1(0

10)1();( 1,0

11

xIppotherwise

or xpppxf xx

xx

X

10 p

pXE ][

qpetm

pqXVart

X

)(

][

04/10/23 4Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 5: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Binomial distribution

• Binomial distribution represents the probability of having exactly x success in n independent and identical Bernoulli trials.

)()1(

0

,,1,0)1(),;( ,,1,0 xIpp

x

n

otherwise

nxppx

npnxf n

xnxxnx

X

npXE ][nt

X peqtm

npqpnpXVar

)()(

)1(][

04/10/23 5Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 6: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Negative binomial distribution• Negative binomial distribution represents the

probability of achieving the r-th success in x independent and identical Bernoulli trials.

• Unlike the binomial distribution for which the number of trials is fixed, the number of successes is fixed and the number of trials varies from experiment to experiment. The negative binomial random variable represents the number of trials needed to achieve the r-th success.

04/10/23 6Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 7: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

,1,;,2,1)1(1

1),;(

rrx rppr

xprxf rrx

X

prXE /][

rtrtX qepetm

prqXVar

)1/()()(

/][ 2

04/10/23 7Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 8: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Geometric distribution

• Geometric distribution represents the probability of obtaining the first success in x independent and identical Bernoulli trials.

,3,2,1)1();( 1 x pppxf xX

pXE /1][

)1/()()(

/][ 2

ttX qepetm

pqXVar

04/10/23 8Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 9: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Hypergeometric distribution

where M is a positive integer, K is a nonnegative integer that is at most M, and n is a positive integer that is at most M.

otherwise

nx for

n

Mxn

KM

x

K

nKMxfX

0

,,1,0),,;(

04/10/23 9Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 10: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

• Let X denote the number of defective products in a sample of size n when sampling without replacement from a box containing M products, K of which are defective.

MnKXE /][

1][

M

nM

M

KM

M

KnXVar

04/10/23 10Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 11: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Poisson distribution

• The Poisson distribution provides a realistic model for many random phenomena for which the number of occurrences within a given scope (time, length, area, volume) is of interest. For example, the number of fatal traffic accidents per day in Taipei, the number of meteorites that collide with a satellite during a single orbit, the number of defects per unit of some material, the number of flaws per unit length of some wire, etc.

04/10/23 11Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 12: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

0

,2,1,0!

);(

x x

exf

x

X

)(!

xIx

e0,1,

x

][XE ][XVar

)1()( te

X etm

04/10/23 12Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 13: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Assume that we are observing the occurrence of certain happening in time, space, region or length. Also assume that there exists a positive quantity which satisfies the following properties:

1.

04/10/23 13Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 14: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

2.

3.

04/10/23 14Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 15: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

04/10/23 15Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

The probability of success (occurrence) in each trial.

Page 16: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

04/10/23 16Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 17: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

04/10/23 17Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 18: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

04/10/23 18Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

,2,1,0!

);(

x x

exf

x

X

Page 19: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

04/10/23 19Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

0

0.2

0.4

0.6

0.8

1

0 5 10 15 20 25 30 35 40 45 50

alpha=0.05 alpha=0.1 alpha=0.2 alpha=0.5

Page 20: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Comparison of Poisson and Binomial distributions

04/10/23 20Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 21: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

• Example Suppose that the average number of telephone calls

arriving at the switchboard of a company is 30 calls per hour.

(1) What is the probability that no calls will arrive in a 3-minute period?

(2) What is the probability that more than five calls will arrive in a 5-minute interval?

Assume that the number of calls arriving during any time period has a Poisson distribution.

04/10/23 21Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 22: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Assuming time is measured in minutes

04/10/23 22Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Poisson distribution is NOT an appropriate choice.

Page 23: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Assuming time is measured in seconds

04/10/23 23Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Poisson distribution is an appropriate choice.

Page 24: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

• The first property provides the basis for transferring the mean rate of occurrence between different observation scales.

• The “small time interval of length h” can be measured in different observation scales.

• represents the time length measured in scale of .

• is the mean rate of occurrence when observation scale is used.

i

i

hhi

i

i

04/10/23 24Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 25: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

• If the first property holds for various observation scales, say , then it implies the probability of exactly one happening in a small time interval h can be approximated by

• The probability of more than one happenings in time interval h is negligible.

12

21

1

21 21

p

hhh

hhh

nn

n n

n ,,1

04/10/23 25Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 26: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

• probability that more than five calls will arrive in a 5-minute interval

• Occurrences of events which can be characterized by the Poisson distribution is known as the Poisson process.

.042021.0

)5()5()5()5()5()5(1 543210

PPPPPP

04/10/23 26Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 27: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Probability density functions of continuous random variables

• Uniform or rectangular distribution• Normal distribution (also known as the Gaussian

distribution)• Exponential distribution (or negative exponential

distribution)• Gamma distribution (Pearson Type III)• Chi-squared distribution• Lognormal distribution

04/10/23 27Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 28: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Uniform or rectangular distribution

)()(

1),;( ],[ xI

abbaxf baX

2/)(][ baXE

tab

eetm

abXVaratbt

X )()(

12/)(][ 2

04/10/23 28Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 29: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

PDF of U(a,b)

04/10/23 29Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 30: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Normal distribution (Gaussian distribution)

2

2

2

1

2

1),;(

x

X exf

][XE

2/

2

22

)(

][tt

X etm

XVar

04/10/23 30Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 31: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

04/10/23 31Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Z

Page 32: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

04/10/23 32Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 33: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

04/10/23 33Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 34: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

04/10/23 34Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Z~N(0,1)X~N(μ1, σ1) Y~N(μ2, σ2)

2

2

1

1

YX

Z

Page 35: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Commonly used values of normal distributions

04/10/23 35Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 36: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Exponential distribution(negative exponential distribution)

.0)();( ),0[ ,xIexf x

X

/1][ XE

t for t

tm

XVar

X )(

/1][ 2

Mean rate of occurrence in a Poisson process.

04/10/23 36Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 37: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

04/10/23 37Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 38: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Gamma distribution

.0,0,)()(

1),;( ),0[

/1

xIex

xf xX

][XE

./1for )1()(

][ 2

tttm

XVar

X

represents the mean rate of occurrence in a Poisson process.

is equivalent to in the exponential density.

/1

/1

04/10/23 38Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 39: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

• The exponential distribution is a special case of gamma distribution with

• The sum of n independent identically distributed exponential random variables with parameter has a gamma distribution with parameters .

.1

/1 and n

04/10/23 39Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 40: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Pearson Type III distribution (PT3)

, and are the mean, standard deviation and skewness coefficient of X, respectively.

It reduces to Gamma distribution if = 0.

xex

xfx

X

,)(

1)(

1

22

04/10/23 40Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 41: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

• The Pearson type III distribution is widely applied in stochastic hydrology.

• Total rainfall depths of storm events can be characterized by the Pearson type III distribution.

• Annual maximum rainfall depths are also often characterized by the Pearson type III or log-Pearson type III distribution.

04/10/23 41Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 42: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Chi-squared distribution

• The chi-squared distribution is a special case of the gamma distribution with

.1,2,k, )(2)2/(2

1);( ),0[

2/1)2/(

xIex

kkxf x

k

X

kXE ][

.2/1)21()(

2][2/

t for ttm

kXVark

X

.2 and 2/ k

04/10/23 42Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 43: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

04/10/23 43Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 44: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Log-Normal DistributionLog-Pearson Type III Distribution (LPT3)

• A random variable X is said to have a log-normal distribution if Log(X) is distributed with a normal density.

• A random variable X is said to have a Log-Pearson type III distribution if Log(X) has a Pearson type III distribution.

04/10/23 44Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 45: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Lognormal distribution

)(2

1),;( ),0(

ln

2

12

2

xIex

xf

x

X

)2/( 2

][ eXE22 222][ eeXVar

04/10/23 45Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 46: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Approximations between random variables

• Approximation of binomial distribution by Poisson distribution

• Approximation of binomial distribution by normal distribution

• Approximation of Poisson distribution by normal distribution

04/10/23 46Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 47: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Approximation of binomial distribution by Poisson distribution

04/10/23 47Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 48: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Approximation of binomial distribution by normal distribution

• Let X have a binomial distribution with parameters n and p. If , then for fixed a<b,

is the cumulative distribution function of the standard normal distribution.

It is equivalent to say that as n approaches infinity X can be approximated by a normal distribution with mean np and variance npq.

n

)()( abnpqbnpXnpqanpPbnpq

npXaP

)(x

04/10/23 48Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 49: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Approximation of Poisson distribution by normal distribution

• Let X have a Poisson distribution with parameter . If , then for fixed a<b

• It is equivalent to say that as approaches infinity X can be approximated by a normal distribution with mean and variance .

)()( abbXaPbX

aP

04/10/23 49Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 50: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

04/10/23 50Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 51: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Example

• Suppose that two fair dice are tossed 600 times. Let X denote the number of times that a total of 7 dots occurs. What is the probability that ?

04/10/23 51Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

11090 X

Page 52: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

04/10/23 52Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 53: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Transformation of random variables

• [Theorem] Let X be a continuous RV with density fx. Let Y=g(X), where g is strictly

monotonic and differentiable. The density for Y, denoted by fY, is given by

.)(

))(()(1

1

dy

ydgygfyf XY

04/10/23 53Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 54: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

• Proof: Assume that Y=g(X) is a strictly monotonic increasing function of X.

04/10/23 54Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 55: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Example• Let X be a gamma random variable with

04/10/23 Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

55

1

1

1

11)(

1

)(

1)(

,,Let

Y

Y

Y

XXX

eY

eY

yf

dY

dXYX

XXY

.0,0,)()(

1),;( ),0[

/1

xIex

xf xX

Y is also a gamma random variable with scale parameter 1/ and shape parameter .

Page 56: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Definition of the location parameter

04/10/23 56Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 57: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Example of location parameter

04/10/23 57Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 58: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Definition of the scale parameter

04/10/23 58Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 59: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Example of scale parameter

04/10/23 59Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 60: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Simulation Simulation • Given a random variable X with CDF FX(x), there

are situations that we want to obtain a set of n random numbers (i.e., a random sample of size n) from FX(.) .

• The advances in computer technology have made it possible to generate such random numbers using computers. The work of this nature is termed “simulation”, or more precisely “stochastic simulation”.

04/10/23 60Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 61: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

04/10/23 61Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 62: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Pseudo-random number generation

• Pseudorandom number generation (PRNG) is the technique of generating a sequence of numbers that appears to be a random sample of random variables uniformly distributed over (0,1).

04/10/23 62Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 63: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

• A commonly applied approach of PRNG starts with an initial seed and the following recursive algorithm (Ross, 2002)

modulo m where a and m are given positive integers, and the

above equation means that is divided by m and the remainder is taken as the value of .

1 nn axx

1naxnx

04/10/23 63Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 64: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

• The quantity is then taken as an approximation to the value of a uniform (0,1) random variable.

• Such algorithm will deterministically generate a sequence of values and repeat itself again and again. Consequently, the constants a and m should be chosen to satisfy the following criteria:– For any initial seed, the resultant sequence has the “appearance” of

being a sequence of independent uniform (0,1) random variables.– For any initial seed, the number of random variables that can be

generated before repetition begins is large.– The values can be computed efficiently on a digital computer.

mxn /

04/10/23 64Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 65: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

• A guideline for selection of a and m is that m be chosen to be a large prime number that can be fitted to the computer word size. For a 32-bit word computer, m = and a = result in desired properties (Ross, 2002).

1231 57

04/10/23 65Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 66: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Simulating a continuous random variable

• probability integral transformation

04/10/23 66Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

Page 67: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

04/10/23 Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

67

The cumulative distribution function of a continuous random variable is a monotonic increasing function.

Page 68: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Example

04/10/23 Lab for Remote Sensing Hydrology and Spatial Modeling Dept of Bioenvironmental Systems Engineering, NTU

68

• Generate a random sample of random variable V which has a uniform density over (0, 1).

• Convert to using the above V-to-X transformation.

)1,0(~,ln)1ln(

1)()(

)(

0

UiidVVU

X

UeduufxF

exf

xx

x

},,,{ 21 nvvv

},,,{ 21 nxxx },,,{ 21 nvvv

Page 69: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Random number generation in R• R commands for stochastic simulation (for

normal distribution – pnorm – cumulative probability– qnorm – quantile function– rnorm – generating a random sample of a specific

sample size– dnorm – probability density function

For other distributions, simply change the distribution names. For examples, (punif, qunif, runif, and dunif) for uniform distribution and (ppois, qpois, rpois, and dpois) for Poisson distribution.

04/10/23 69Lab for Remote Sensing Hydrology and Spatial Modeling Dept of Bioenvironmental Systems Engineering, NTU

Page 70: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

Generating random numbers of discrete distribution in R

• Discrete uniform distribution– R does not provide default functions for random

number generation for the discrete uniform distribution.

– However, the following functions can be used for discrete uniform distribution between 1 and k.• rdu<-function(n,k) sample(1:k,n,replace=T) # random number• ddu<-function(x,k) ifelse(x>=1 & x<=k & round(x)==x,1/k,0) # density• pdu<-function(x,k) ifelse(x<1,0,ifelse(x<=k,floor(x)/k,1)) # CDF• qdu <- function(p, k) ifelse(p <= 0 | p > 1, return("undefined"),

ceiling(p*k)) # quantile

04/10/23 Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

70

Page 71: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

– Similar, yet more flexible, functions are defined as follows• dunifdisc<-function(x, min=0, max=1) ifelse(x>=min & x<=max &

round(x)==x, 1/(max-min+1), 0)>dunifdisc(23,21,40)>dunifdisc(c(0,1))

• punifdisc<-function(q, min=0, max=1) ifelse(q<min, 0, ifelse(q>max, 1, floor(q-min+1)/(max-min+1)))>punifdisc(0.2)>punifdisc(5,2,19)

• qunifdisc<-function(p, min=0, max=1) floor(p*(max-min+1))+min>qunifdisc(0.2222222,2,19)>qunifdisc(0.2)

• runifdisc<-function(n, min=0, max=1) sample(min:max, n, replace=T)>runifdisc(30,2,19)>runifdisc(30)

04/10/23 Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

71

Page 72: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

• Binomial distribution

04/10/23 Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

72

Page 73: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

• Negative binomial distribution

04/10/23 Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

73

Page 74: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

• Geometric distribution

04/10/23 Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

74

Page 75: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

• Hypergeometric distribution

04/10/23 Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

75

Page 76: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

• Poisson distribution

04/10/23 Laboratory for Remote Sensing Hydrology and Spatial Modeling, Dept of Bioenvironmental Systems Engineering, National Taiwan Univ.

76

Page 77: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

An example of stochastic simulation

• The travel time from your home (or dormitory) to NTU campus may involve a few factors:– Walking to bus stop (stop for traffic lights,

crowdedness on the streets, etc.)– Transportation by bus– Stop by 7-11 or Starbucks for breakfast (long queue)– Walking to campus

04/10/23 77Lab for Remote Sensing Hydrology and Spatial Modeling Dept of Bioenvironmental Systems Engineering, NTU

Page 78: STATISTICS Univariate Distributions Professor Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering National Taiwan University

• If you leave home at 8:00 a.m. for a class session of 9:10, what is the probability of being late for the class?

)36,15(~ 21 NX

~2X Gamma distribution with mean 30 minutes and standard deviation 10 minutes.

~3X Exponential distribution with a mean of 20 minutes.

)25,10(~ 24 NX All Xi’s are independently distributed.

04/10/23 78Lab for Remote Sensing Hydrology and Spatial Modeling Dept of Bioenvironmental Systems Engineering, NTU

4321 XXXXY