Presentation is loading. Please wait.

Presentation is loading. Please wait.

1 Discrete Distributions Ch3. 2   Def.3.1-1: A function X that assigns to each element s in S exactly one real number X(s)=x is called a random variable.

Similar presentations


Presentation on theme: "1 Discrete Distributions Ch3. 2   Def.3.1-1: A function X that assigns to each element s in S exactly one real number X(s)=x is called a random variable."— Presentation transcript:

1 1 Discrete Distributions Ch3

2 2   Def.3.1-1: A function X that assigns to each element s in S exactly one real number X(s)=x is called a random variable. The space of X is the set of real numbers {x: X(s)=x, s ∈ S}.   If X(s)=s, then the space of X is also S.   Ex3.1-2: S={1,2,3,4,5,6} in casting a dice. Let X(s)=s.   P(X=5)=1/6, P(2≤X≤5)=P(X=2,3,4,5}=4/6, P(X≤2)=P(X=1,2)=2/6.   Two major difficulties:   In many practical situations, the probabilities assigned to the events are unknown.   Repeated observations (Sampling) to estimate them.   There are many ways to define X. Which to use?   Measurement, materialization of outcomes.   To draw conclusions or make predictions.

3 3 Random Variables of the Discrete Type   When S contains a countable number of points,   X can be defined to correspond each point in S to a positive integer.   S is a set of discrete points or a discrete outcome space.   X is a random variable of discrete type.   The probability mass function (p.m.f.) f(x) denotes P(X=x).   f(x) is also called probability function, probability density function, frequency function.   Def.3.1-2: f(x) of a discrete random variable X is a function:   f(x)>0, x ∈ S.   ∑ x ∈ S f(x) = 1,   P(X ∈ A) = ∑ x ∈ A f(x), where A ⊂ S.   f(x)=0 if x ∉ S: S is referred to as the support of X and the space of X.   A distribution is uniform if its p.m.f. is constant over the space.   For instance, f(x)=1/6 in rolling a fair 6-sided dice.   Ex.3.1-3: Roll a 4-sided die twice and let X be the largest of two outcomes. S={(i, j): i, j=1..4}. P(X=1)=P({(1,1)})=1/16. P(X=2)=P({(1,2), (2,1), (2,2)})=3/16, …, f(x)=P(X=x)=(2x-1)/16 for x=1..4; f(x)=0 otherwise.

4 4 Graphing the Distribution   (a) For a discrete probability distribution, we simply plot the points (x, f(x)), for all x  R.   (b) To get a better picture of the distribution, we use bar graphs and histograms.   (c) A bar graph is simply a set of lines connecting (x, 0) and (x, f(x)).   (d) If X takes on only integer values, then a histogram can be used by using rectangles centered at each x  R of height f(x) and width one.

5 5 Graphic Representation of f(x)   The graph consists of the set of points {(x, f(x): x ∈ S}.   Better visual appreciation: a bar graph, a probability histogram.   P(X=x)=f(x)=(2x-1)/16, x=1..4.   Hyper-geometric distribution: a collection has N 1 objects of type 1 and N 2 objects of type 2. X is the number of type-1 objects in the n objects taken from the collection w/o replacement.

6 6 Examples   Ex.3.1-5: In a pond of 50 fish with 10 tagged, 7 fish are caught at random w/o replacement.   The probability of 2 tagged fish caught is   Ex.3.1-7: In a lot of 25 items with unknown defective, 5 items are selected at random w/o replacement for exam.   If no defective found, the lot is accepted; otherwise, rejected.   Given N 1 defective, the acceptance probability, operating characteristic curve, is   Ex.3.1-8: Roll a 4-sided die twice. X is the sum: 2..8.   f(x)=(4-|x-5|)/16.   1000 experiments simulated on a computer.

7 7 Mathematical Expectation   Def.3.2-1: For a random variable X with p.m.f. f(x), if ∑ x ∈ S u(x)f(x) exists, it is called the mathematical expectation or the expected value of the function u(X), denoted by E[u(X)].   It is the weighted mean of u(x), x ∈ S.   The function u(X) is also a random variable, say Y, with p.m.f. g(y).   Ex.3.2-2: For a random variable X with f(x)=1/3, x ∈ S={-1, 0, 1}.   Let u(X)=X 2. Then, E[u(X)]= E[X 2 ]=2/3.   The support of the random variable Y=X 2 is S 1 ={0, 1}, and P(Y=0)=P(X=0), P(Y=1)=P(X=-1)+P(X=1), so its p.m.f.   Hence ∑ y ∈ S1 yg(y) = 2/3, too.   Thm.3.2-1: mathematical expectation E, if exists, satisfies:   For a constant c, E(c)=c.   For a constant c & a function u, E[c u(X)]=c E[u(X)].   For constants a & b and functions u & v, E[a u(X) + b v(X)]=a E[u(X)] + b E[v(X)].   It can be applied for 2+terms since E is a linear or distributive operator.

8 8 Examples   Ex3.2-3: f(x)=x/10, x=1,2,3,4.   Ex3.2-4: u(x)=(x-b) 2, where b is a constant.   Suppose E[(X-b) 2 ] exists. What is the value of b to minimize it?   Ex3.2-5: X has a hypergeometric distribution.

9 9 More Examples   Ex.3.2-6: f(x)=x/6, x=1,2,3.   E(X)=μ=1(1/6)+2(1/6)+3(1/6)=7/3.   Var(X)=σ 2 =E[(X-μ) 2 ]=E(X 2 )-μ 2 =…=5/9 ⇒ σ=0.745   Ex.3.2-7: f(x)=1/3, x=-1,0,1.   E(X)=μ=-1(1/3)+0(1/3)+1(1/3)=0.   Var(X)=σ 2 =E(X 2 )-μ 2 =…=2/3 ⇒ The standard deviation σ=0.816   Comparatively, g(y)=1/3, y=-2,0,2.   Its mean is also zero; but, Var(Y)=8/3 and σ Y =2σ. ⇒ more spread out.   Ex.3.2-8: uniform f(x)=1/m, x=1..m.   E(X)=μ=1(1/m)+…+m(1/m)=(m+1)/2.   Var(X)=σ 2 =E(X 2 )-μ 2 =…=(m2-1)/12   For instance, m=6 when rolling a 6-sided die.

10 10 Derived Random Variables   Linear Combination: X has a mean μ X and variance σ X 2.   Y=aX+b ⇒ μ Y = aμ X +b; Var(Y)=E[(Y-μ Y ) 2 ] =…=a 2 σ X 2 ; σ X =|a|σ X.   a=2, b=0 ⇒ mean*2, variance*4, standard deviation*2.   a=1, b=-1 ⇒ mean-1, variance*1, standard deviation*1. Var(X-1) = Var(X).   The r th moment of the distribution about b: E[(X-b) r ].   The r th factorial moment: E[(X) r ]=E[X(X-1)(X-2)…(X-r+1)].   E[(X) 2 ] = E[X(X-1)] = E(X 2 )-E(X) = E(X 2 )-μ.   E[(X) 2 ]+μ-μ 2 = E(X 2 )-μ+μ-μ 2 = E(X 2 )-μ 2 = Var(X)=σ 2.   Ex3.2-9: X has a hypergeometric distribution. (ref. Ex3.2-5)

11 11 Bernoulli Trials   A Bernoulli experiment is a random experiment, whose outcome can be classified in one of two mutually exclusive and exhaustive ways: success or failure.   A series of Bernoulli trials occurs after independent experiments.   Probabilities of success p and failure q remain the same. (p+q=1)   Random variable X follows a Bernoulli distribution.   X(success)=1 and X(failure)=0.   The p.m.f. of X is f(x)=p x q (1-x), x=0,1.   (μ, σ 2 )=(p, pq).   A series of n Bernoulli trials, a random sample, will be an n-tuple of 0/1’s.   Ex.3.3-4: Plant 5 seeds and observe the outcome (1,0,1,0,1): 1st, 3rd, 5th seeds germinated. If the germination probability is.8, the probability of this outcome is (.8)(.2)(.8)(.2)(.8) assuming independence.   Let X be the number of successes in n trials.   X follows a binomial distribution, denoted as b(n, p).   The p.m.f. of X is

12 12 Example   Ex.3.3-5: For lottery with.2 winning, if X equals the number of winning tickets among n=8 purchases.   The probability of having 2 winning tickets is   Ex.3.3-6: The effect of n and p is illustrated as follows.

13 13 Cumulative Distribution Function   The cumulative probability F(x), defined as P(X≤ x), is called the cumulative distribution function or the distribution function.   Ex.3.3-7: Assume the distribution of X is b(10, 0.8).   F(8) = P(X≤8) = 1-P(X=9)-P(X=10) = 1-10(.8) 9 (.2)-(.8) 10 =.6242.   F(6) = P(X≤6) = ∑ x=0..6 C x 10 (.8) x (.2) 10-x.   Ex.3.3-9: Y follows b(8, 0.65).   If X=8-Y, X has b(8,.35), whose distribution function is in Table II (p.647).   E.g., P(Y≥6) = P(8-Y≤8-6) = P(X≤2)=0.4278 from table lookup.   Likewise, P(Y≤5) = P(8-Y≥8-5) = P(X≥3) = 1-P(X≤2) = 1-0.4278 =0.5722.   P(Y=5) = P(X=3) = P(X≤3)-P(X≤2) = 0.7064-0.4278 = 0.2786.   The mean and variance of the binomial distribution is (μ, σ 2 )=(np, npq).   Ex.3.4-2 details the computations.

14 14 Comparisons   Empirical Data vs. Analytical Formula :   Ex.3.3-11: b(5,.5) has μ= np= 2.5, σ 2 = npq= 1.25.   Simulate the model for 100 times: 2, 3, 2, … ⇒ = 2.47, s 2 =1.5243.   Suppose an urn has N1success balls and N 2 failure balls. N = N 1 + N 2.   Let p = N 1 /N, and X be the number of success balls in a random sample of size n taken from this urn.   If the sampling is done one at a time with replacement, X follows b(n, p).   If the sampling is done without replacement, X has a hypergeometric distribution with p.m.f.   If N is large and n is relative small, it makes little difference if the sampling is done with or without replacement. (See Fig.3.3-4)

15 15 Moment-Generating Function (m.g.f )   Def.3.4-1: X is a random variable of the discrete type with p.m.f. f(x) and space S. If there is a positive integer h s.t.   E(e tX ) = ∑ x ∈ S e tx f(x) exists and is finite for –h<t<h, then M(t) = E(e tX ) is called the moment-generating function of X.   –E(e tX ) exists and is finite for –h<t<h ⇔ M (r) (t) exist at t=0, r=1,2,3,…   –Unique association: p.m.f. ⇔ m.g.f.   Sharing the same m.g.f., two random variables have the same distribution of probability.   Ex.3.4-1: X has m.g.f M(t) = e t (3/6)+ e 2t (2/6)+ e 3t (1/6).   From e“ x”t, Its p.m.f. has f(0)=0, f(1)=3/6, f(2)=2/6, f(3)=1/6, f(4)=0, …   Therefore f(x)=(4-x)/6, x=1,2,3..   Ex.3.4-2: X has m.g.f M(t) = e t /2(1-e t /2), t<ln2.   (1-z) -1 = 1 + z + z 2 + z 3 + …, |z|<1.

16 16 Application of Application of m.g.f   M(t) = E(e tX ) = ∑ x ∈ S e tx f(x) exists and is finite for –h<t<h. M'(t) = ∑ x ∈ S xe tx f(x), M'(0) = ∑ x ∈ S xf(x) = E(X), M''(t) = ∑ x ∈ S x 2 e tx f(x), …, M''(0) = ∑ x ∈ S x 2 f(x) = E(X 2 ), …, M (r) (t) = ∑ x ∈ S x r e tx f(x). M(r)(0) = ∑ x ∈ S x r f(x) = E(X r ), as t=0.   M(t) must be formulated (in closed form) to get its derivatives of higher order.   Ex.3.4-3: X has a binomial distribution b(n, p).   Thus, its m.g.f. is   When n=1, X has a Bernoulli distribution.

17 17 Negative Binomial Distribution   Let X be the number of Bernoulli trials to observe the r th success.   X has a negative binomial distribution.   Its p.m.f. g(x) is   If r=1, X has a geometric distribution with

18 18 Geometric Distribution   X has a geometric distribution with the p.m.f.   P(X > k), P(X ≤k):   Memory-less:   (EX3.412)   Ex3.4-4: Fruit flies’ eyes with ¼white and ¾red.   The probability of checking at least 4 flies to observe a white eye is P(X≥4) = P(X>3) = (¾) 3 = 0.4219.   The probability of checking at most 4 flies to observe a white eye isP(X≤4) = 1-(¾) 3 =0.6836.   The probability of finding the first white eye on the 4th fly checked is P(X=4) = pq 4-1 = 0.1055. <= P(X≤4) -P(X≤3)   Ex3.4-4: For a basketball player with 80% free throw.   X is the minimum number of throws for a total of 10 free throws.   Its p.m.f. is   μ= r/p= 10/0.8 = 12.5, σ 2 = rq/p 2 = 10(0.2)/(0.8) 2 = 3.125.

19 19 m.g.f.. ⇒ p.d.f   By Maclaurin’sseries expansion (ref. p.632): If the moments of X, E(X r ) = M (r) (0), are known, M(t) is thus determined. ⇒ p.d.f. can be obtained by rewriting M(t) as the weighted sum of e “x” t.   Ex3.4-7: If the moments of X are E(Xr) = 0.8, r=1,2,3,…   Then, M(t) can be determined as Therefore, P(X=0)=0.2 and P(X=1)=0.8

20 20 Poisson Process   Def.3.5-1: An approximate Poisson process with parameter λ>0:   The numbers of changes occurring in non-overlapping intervals are independent.   The probability of exactly one change in a sufficiently short interval of length h is approximately λh.   The probability of two or more changes in a sufficiently short interval is essentially zero.   Determine p.m.f.:   During the unit interval of length 1, there are x changes.   For n»x, we partition the unit interval into n subintervals of length 1/n.   The probability of x changes in the unit interval ≡The probability of one change in each of exactly x of these n subintervals.   The probability of one change in each subinterval is roughly λ(1/n).   The probability of two or more changes in each subinterval is essentially 0.   The change occurrence or not in each subinterval becomes a Bernoulli trial.   Thus, for a sequence of n Bernoulli trials with probability p = λ/n, P(X=x) can be approximated by (binomial):

21 21 Poisson Distribution

22 22 Examples  Ex.3.5-1: X has a Poisson distribution with a mean of λ=5.  Table III on p. 652 lists selected values of the distribution.  P(X≤6) = 0.762  P(X>5) = 1-P(X≤5) = 1-0.616 =0.384  P(X=6) = P(X≤6)-P(X≤5) = 0.762-0.616 = 0.146  Ex.3.5-2: The Poisson probability histograms:

23 23 More Examples  Empirical data vs. Theoretical formula (Ex.3.5-3)  X is the number of αparticles emitted by barium-133 in.1 sec and counted by a Geiger counter.  100 observations are made.  Generally with unit interval of length t, the Poisson p.m.f. is  Ex.3.5-4: Assume a tape flaw is a Poisson distribution with a mean of λ= 1/1200 flaw per feet.  What is the distribution of X, the number of flaws in a 4800-foot roll?  E(X)=4800/1200=4.  P(X=0) = e -4 = 0.018  By Table III on p. 652, P(X≤4) =0.629

24 24  Poisson with λ=np can simulate Binomial for large n and small p.  (μ, σ 2 )=(λ, λ) ≈(np, npq).  Ex3.5-6: Bulbs with 2% defective rate.  The probability that a box of 100 bulbs contains at most 3 defective bulbs is  By Binomial, the tedious computation will result in  Ex3.5-7: p.160, the comparisons of Binomial and Poisson distributions. When Poisson ≈ Binomial

25 25  Ex3.5-8: Among a lot of 1000 parts, n=100 parts are taken at random w/o replacement. The lot is accepted if no more than 2 of100 parts taken are defective. Assume p is the defective rate.  Operating characteristic curve OC(p) = P(X≤2) = P(X=0)+P(X=1)+P(X=2). Hypergeometric: N 1 =1000p, N=1000, N 2 =N-N 1  Since N is large, it makes little difference if sampling is done with or without replacement. ⇒ simulated by Binomial!  n Bernoulli trials:  For small p, p=0~.1: ⇒ simulated by Poisson!  OC(0.01)=0.92 OC(0.02)=0.677 OC(0.03)=0.423 OC(0.05)=0.125 OC(0.10)=0.003 When Poisson ≈ Binomial Binomial ≈ Hypergeometric Hypergeometric


Download ppt "1 Discrete Distributions Ch3. 2   Def.3.1-1: A function X that assigns to each element s in S exactly one real number X(s)=x is called a random variable."

Similar presentations


Ads by Google