ete 1.4 Binomial Distribution Consider an experiment with the following properties. ie., trials ed number n of Bernoulli’ trials, 1. The experiment consists of a fix failure (f)- such that the result is either success (s) or erefore, the probability of independent and the e from trial to trial. mber of successes obtaine: 2. The trials are identical and success ‘p” remains the sami fori trie 3, The random variable X denotes the nu ae als. To understand the binomial density let us consider the situation for.n = 3. The sample space of such an experiment is SHUTS. sff. fof. ffs, 98h, 9f% fiss, sss} Let us consider the probabilities ofthe sample points f ff and sf f. PUff)=(—p)t—P)-P)= (1p) (the trials are independent). f) = p(t p)(1 7) = Pld pyRandom Variables 1,105 The random variable X takes the value 0 if the experiment results in an outcome SEF ie P(X = 0) = (1 p)# But the random variable X assumes the value 1 in any one of the outcome ff. fst, fss. Hence = P(X = 1) = 3p(1— p)? Similarly P(X = 2) = 3p(1 — p) PU p 0,1, 2,3. Hence for P(X = 2) = e(x)p*(1 - p)~* where:c(z) denotes the number of sample points that correspond to x successes. These sample points are expressed as permutations of three letters with x of these being s and the rest 3 — x being f. 3! Hence (x) = ae = 3cz Thus the probability function of the binomial variable is given by p(x) = 3¢, p?(1— p)3"*_ x =0,1,2,3. Generalizing for n trials =0,1,2,3,....7 P(z) = ne,p*(1 —p)"*- Definition Arandom variable X is said to follow a binomial distribution if it assumes only non-negative values with probability mass function. P(X-= 2) = ne,p*q™ where x = 0,1,2,3,...,n. and qg=1—p io the two indepéndent constants n and p are the parameters of binomial distri- ion. * :1.106 Probability and Random Processes 1.4.1. Moments of binomial distribution about the origin ‘The p.m.f of Binomial distribution is P(X =a)= nea” oF 0,1, 243) rer Now pre (X)= 2) fo = x rnc" i . ye ae art goign) = % Dewan? y binomial theorem) -r = nota)" 1 @ == np (Since p+q= 1) ~ Hence the mean of the Binomial distribution is np. = E(X?) = 32292) z=0 n =yo2 sere = Dle(e- 1) +2) (Cz) P*g"* = = Soale = 1) "Czp*q"* + ys "C,pFq?-* z=0 z=0 = n(n - —1)p? > (2. E,_qypP2qi"-2)-(2—) + np ‘222 > = n(n 1)p°(q + p)"? + np nln pt + np (. p+q=1) i va EX?) + [E(X)}? np?(n,—-l) + np — n2p?Random Variables’ 1.107 = np? — np? + np — n2p? np(1 — p) = npq Var(X) = npq 13 = E[X®] =) x4p(z) 2=0 = Solel - 1)(@ — 2) + 82(a = 1) +2] "Cp" z=0 Hg = n(n — 1)(n — 2)p? + 8n(n — 1)p? + np cea 4 = E[X4] = > x4p(z) io = Slee = 1(@ — 2)(e— 3) + 6x(e — 1)(e = 2) + 7e(e -1) +a] 2=0 "Cpt Ha = n(n — 1)(n — 2)(n — 3)p* + 6n(n — 1)(n — 2)p* + Tn(n — 1)p? + np 1.4.2 Moment Generating Function (M.G.F) (AU, May/June 2006, MA034, April/May 2004,’ MA040) (AU, Nov/Dec-2008, MA 1255) (May/June 2009, MA 1040) Mx(t) = E(e*) = Se'*p(x) Ford Soe Cypeg* 2=0" n => "Cx(etp)Fg"* z=0 Mx (t) = (pet +4)” . Note Mean and Variance using M.G.F t=0 Mean = [fx] = [n(pet + q)”"'pe'],_1.108 Probability and Random Processes = [npet(pe! +4)""]ino =nplp+q)t=np [2 ptT™ y . Mean = np [ © (pet + a] oS E(x?) = [Sato] wo Le dt? 7 [jinn oe +0}, = np[e'(n—- 1)(pet +." Pe" + (eet +9)" 220 = np[(n— Detar P+ @+ a) = npl(n — 1p + = nein — P+ 1) = np = np? + pq Variance = E(X?] ~ (EC)? = n®p? + npq— (np)? np +4) Distribution dent random variables, then s ‘Additive Property of Binot 1 X ~ B(m,p) and Y ~ B(nz, p) are indepen X+Y¥ ~ Bim +n?) Mx (t) = (9+ pe')™ My (t) = (q+ pe')”? j X and Y are independent | Mxsy(t) = Mx(t).My (t) | M ‘ = (qt pet)".(q-+ pe)" | he! (qt pein which is the m.g-f of a binomial variate with parameters (m1 + 72, P)Random Variables 1.109 Solved Problem 1.84 For a Binomial distribution of mean 4 and variance 2, find the probability of get- ting (i) atleast 2 successes. (ii) atmost 2 successes (iii) Find P(5 < X <7) Solution The P.m.f. of Binomial distribution is P(e) =" Carat Given mean. = np =4 variance = npq = 2 > Ne ow np 4 Fag 1 1 Hence p=1-q=5 Totetns pet = net 2 n=8 1 1 P=p Vg r=8 Cs. * P(X =2)="C, (ay Q). (i) Pat least 2 successes) = P(X > 2) =1-P(X <2) =1- [P(X =0)+ P(X =1)] = 1 = PCop%g?? + 8Crp'g®1] =1-[(8)'+8() (2) [QQ] a2 ()'a+8 7 1- 3(9) 2 2a 256 «256 =1-L110 Probubility and Random Processes (ii) P(atmost 2 successes) = P(X < 2) = P(X =0)+ P(X = 1) + P(X = 2) = qh +8Cipg + 5Cop?a® -()'@ @ +) @ 8 =(1+8+28) (3) 2 He “aa(s)' Ge ()'(@)" it) PO SKS Ived Problem 1.85 AU, April / May 2008, MA i283 ‘A Binomial variable X satisfies the relation 9 P(X = 4) = P(X = 2) wher 6. Find the parameter p of the Binomial distribution. Solution i The Probability function for a Binomial distribution is P(X = 2) ="Czp"q"* eee d= 6Cyptg? since n =6 “P(X =2) = "Capa! | Given ap fro =2) | 9:5Cyp'q? = SCop%q! \ | 1385p? = 159? 9p? —q? =0 ZaRandom Variables 1.111 op? = (1 = p)? Op? — (1 + p* — 2p) =0 Op? —1—p? + 2p =0 8p? + 2p-1=0 ~24 JI=(A(B-T) _ -24 VIF BZ 16 ie To 2+£6 -1+3 p= p cannot be negative Solved Problem 1.86 (AU, May/June 2009, MA 1259) _If, on an average, 9 ships out of 10 arrive safely to a port, obtain the mean and > 10 oo | P= 70 faa n= 150 Mean = np = 150 x a = 135 Variance = npq = 160 x x 7 a = 135 Standard deviation = /npg = V13.5 = on Solved Problem 1.87 A gun is fixed at a target from a certain distance. The Probability of hitting the target is 0.2. Totally two bombs are enough to destroy the target. If six shells are aimed at the target, find the probability of the target being destroyed.1112 Probability and Random Processes Solution Let p be the probability of shell hitting the target. Here p=0.2 > q = 0.8 n=6 Let X be the number of shells required to destroy the target, P(X 22) =1- P(X <2) =1-[P(X =0)+(X = 1) = 1 = [®Co(0.2)°(0.8)° + °C (0.2)! (0.4) 1 — (0.2621 + 0.3932) = 0.3447. Solved Problem 1.88 - A family has four children. Assuming the probability of a male birth j: is! =. 3 find B gpprebability hat there will be X and Y denote the number ee boys and girls in the family. Pp b P(Y)=4 P(X 21) =1- P(x <0)Random Variables ANB Solved Problem 1.89 The overall percentage of failure i ability that out of a group of 6c: Solution 4 Certain examination is 40, What in the prob: didates at least 4 passed the examination. Let p be the probability of passing the examination, Then q = 1 ~ pis the probability of failure 40. 2 Given n = 6 P (at least 4 persons passed the examination) = P(X = 4) + P(X =5) + P(X =6) 88) @)" 8G) Qa) @)" = 0.54432 Solved Problem 1.90 Ina precision bombing attack, the chance of one bomb striking the target is 50%, For a particular target two direct hits are Tequired to destroy it completely. For giv ‘he probability of a bomb striking the target = Let n be the number of bombs obability at least 0.99 Let_X be the number of bombs The probability that out of n bo p= 50% =1 to be dropped’ to destroy completely, with (out of n) that hit the target mbs, at least 2 hit the target. > 0.99 P(X > 2) > 0.99 ie, [1—- P(X <2)} > 0.99 1— [P(X =0) + P(X = 1)] > 0.99 Now, P(X =x) ="C,prqn-2 fs) @a@ soe\ S 1.114 Probability and Random Proce: 1 [Gem 1- [u +1) (3) | 0.99 W len 2p \ A zitn 9" 2 100 + 1000 Jity is satisfied when m= 11. to destroy the target. By trial we can find that the above inequal Hence at least J] bombs are required Solved Problem 1.91 are injected with a drug that inhibits bod search, it was found that the probability of a rat dying from, 0.2. If 10 rats are used, z In a research, rats ly synthesis of prot. By the previous res drug before the experiment is over is 1. How many are expected to die before the experiment ends. 2. What is the probability that at least eight will survive. Solution . Let X be the random variable representing the number of rats that die in & _ experiment. n=10 Here, p=0.2.>q9= 0.8, 1. Expected number of rats that die before the experiment = np = 10 x 0.2=1 P (at least eight will survive) = P (at most 2 will die) = P(X <2) = P(X =0) 4 P(X =1) + P(X =2) = °C 9(0.2)°(0.8)" + 19°C (0.2)"(0.8)" + 10¢,(0.2)°(0.8)° 0.6778 : Solved Problem 1.92 (AU, May/June 2009, MA 1? » ps been found that 80% of printers used on home computers operate cone the time of installation. A particular dealer sells 10 units during a given moRandom Variables 1.115 1, Find the probability that at least nine printers operate correctly on installa- tion. 2. Consider 5 months, in which 10 units are sold per month. What is the probability that at least nine units Operate correctly in each of the 5 months. Solution Let X’ be the random variable representing the number of printers working cor- rectly. ; 80 : P= 399 =0849=02. n=i0 1. P (at least nine printers operate correctly on installation) = P(X 29) = P(X =9) + P(X = 10) = 7°09(0.8)9(0.2) + !°C40(0.8)1(0.2)° = 0.3758 2, Probability that nine units work correctly in a month = 0.3758. The probability that nine units work cotrectly in each of 5 months = (0.3758)° ; = 0.0975 -d Problem 1.93 owner of a lodge with 5 cabins is considering buying air coolers on line to there cabins, He expects that about half of his customers would be willing to rent coolers and finally he buys 3 coolers. Assuming 100% occupancy at all times, Find the probability that 1. There will be more requests for coolers. 2. A customer who requests will get one. Solution Let X be the random variable representing the number of customers requesting for coolers. Given that half of the customers may be willing to rent coolers Nine1.116 Probability and Random Processes @ P(X > 3) = P(X =4) + P(X =5) = 5Cyptge4 + 5Csp°q?® (2) G)+@) =@+0(5) =6(3) =o (ii) P(customer who gets one) = P(X <3) =1- P(X > 3) = 1-0.1875 = 0.8125 Solved Problem 1.94 Accompany produces screws. It is known that 0.01 of total production is defecs, The company sells the screws in packages of 10 and the company annou;, that money will be given back if at least 2 of the 10 screws is defective, yy, proportion of the packages sold must be replaced? Use binomial distribution, Solution If X be the number of defective screws in the package "The probability that a screw is defective =p =0.01 => q=0.99 The probability that the package will have to be replaced. P (at least two out of 10 screws is defective) HP(X> 1)=1-P(X <1) — (P(X = 0) + P(X = 1)) =1-— (°Cop?g'?- 0 + 100, pig) p) = 1 - (9€9(0.01)°(0.99)° + °c, (0.01)(0.99)") = 1 = ((0.99)!° + 10(0.01)(0.99)") =0.0052,Standom Variables VAST Solved Problem 1.95 ‘The screws produced by a certain machine were checked by examining sarnples of 7. The following table shows the distribution of 12% samples according to the number of defective items they contain Number of defectives oO 1 2 2 4 5 6 T Total ina sample of 7 Number of samples : 7 6 J9 35 30 23 7 J 124 Fit a binomial distribution and find the expected frequencies if the chance of the machine being defective is }. Solution Let X be the number of defective items ir Given p=1/2, n 4 The binomial distribution is given by a sample of 7. 1/2 PIX = "Cy yl? 5 2 = 0,1, 2-07 2) 0)" 0,1,2,---7 7 Expected frequency of getting x defectives = 128 x 7C, ( 3) = 7C,; ©=0,1,2--- The following table gives the expected frequency No. ofdefectives : 0 1 2 3 4 5 67 1 1 Observed frequency : 7 6 19 35 30 23 7 Expected frequency : 1 7 21 35 35 21 7 Solved Problem 1.96 ~ AU, May/June 2009, MA 1252 There is rainfall in a’certain place for 10 days in every thirty days. Find the probability that 1, There is rainfall on at least 3 days of a given week. 2. The first four days of a given week will be wet and the remaining days dry1.118 Probability and Random Processes Solution Let X be the number of rainy days inn week, Given the probability of rainfall i 1,22 -1 Pega Gt : P(X =a) = TCep"g 1. P(there is rainfall on at least 3 days) P(X > 3) =1- P(X <3) 1 = [P(X = 0) + P(X = 1) + P(X = 2)) -- fay G) va) Gv a MOT. [9'"O0'"O'O] = 1-0.5705 = 0.4295 2 P(w we ws wa dy dz dg) = P(w1)P(w2) P(ws)P (wa) P(di)P (da) P(ds) = Pw)/(P(@p- -()'@) 0.0037Random Variables 1.119 Let n be the number of trials. P (at least one success) = P(X > 1) =1-P(X <0) 1—"Cop°g"? =1-q"=1-(0.98)" i 1 that the probability of at least one success is greater than 3 ie, 1—(0.98)" > 5 ee (0.98)" <= log on both sides 1og(0.98)" < log(0.5) nlog(0.98) < log(0.5) n(—0.0087) < 0.3010 0.3010 n> 0.0087 = 34.59 n=35 Problem 1.98 (AU Nov/Dec 2003, MA040) -are thrown 729 times. How many times do you expect at least three dice to va five or a six? Success = getting 5 or 6 with 1 dice Pp = Probability of getting 5 or 6 with one dice Tet X be a random variable denoting the number of successes when 6 dice are thrown once --1.120° Probability and Random Processes Given n=6 then P(X = 2) =°Cs (3) (3)sn P (at least 3 dice show five or six) = P(X D3) = P(X =F P(X = DT P(X =) + PX = 6) af Qa G” al eG = #2002)° + (15)(2)? + (6)(2) + 1] 1 160-460 + 12-41) = 282 = 2 = 3g l160 + 3° 729 6 dice are thrown 729 times. Hence, the expected number of times at least thre, dice show fivé or six ee 233 re 729 Solved Problem 1.99 10 coiris are thrown simultaneously. Find the probability of getting atleast 7 heads Solution Here success is getting a head. Let X denote the numbers of heads obtained. Here X follows binomial dis tribution with 1 : =10, p== and g== a ae. 2 = . P(X =2)= "Opp*gh* 1)" /1\ 0-2 a 1\10 = °C, (3) (3) = 0G, (3) 2 =0,1,2+" Pfatleast 7 heads] =P[X>7] = PIX = 71+ PIX =8]+ PIX =9]4 PIX = 10) - = 8ar(3) "+ 1, 1) yf) 1) \2 Ce 3) + "Co (3) + cio (5)Random Variables 1.121 uo = () [Cr + Cy + Cy + Cio) 1 = 176 = pi 120 +45 + 10+ 1] = 7 = 0.1718 Solved Problem 1.100 A radar system has a probability of 0.1 of detecting a certain target during a single scan. Find the probability that the target will be detected - 1, At least 2 times in four consecutive scans (event A) 2. At least once in twenty scans (event B) Solution Let P (target is detected in single scan) = 0.1 @ P(A) = 4Cop?q4-? + 4Cp8qt-3 4. 4Oyptgt—4 z 2 3 4 ~°(35) (a0) +4(G5) (a0) + Gn) = e = 0.0523 Gi) P(B) = 1 — Piero detection] 20 va =1-g%=1- (3) =1-0.1216 = 0.8784 Solved Problem 1.101 Let X,Y be independent random variables with Binomial distribution. Let X and Y have n = 3,p = d and n= 5, p = 3 Fespectively. Find the probability of rt+y>1. Solution - : Le X~B (3, ) y~B(s, 5)1.122 Probability and Random Processes X and ¥ are independent Binomial random variables with probability of Me, of both X and ¥ equal to }. E Hence by additive property, we get = s+y~a (2452) 2+0~5(83) pety=n="e-(3), ol =1-P(rt+y<1) =1-P(x+y=0) (a) @") Hence P(r+y21) 7 8 ie, P(r+y>1)=1- (3) = 0.961 Solved Problem 1.102 9 The MGF ofa rv X is 3 + 3") show that * 5 7 a J 2 9-2 Pip 20 >, (1)"(2 2% (3) G zal . - Hence proved. Solved Problem 1.103 : ___AU, May / June 2008, MA 1252 A discrete R.V X has moment generating function Mx-(t) = (1 + 3 et)® Find E(X), Var (X) and P(X = 2) Solution . The given m.g.f is of the form (q + pet)" + 3et)° -. X follows binomial distribution with . 3 1 PHD a7 n=5 5- 5 Gee ENA P(X =2)= 5c; A) (; +92 =0,1,2,3, 4,5, raanara()'() * = 00 45 ~ 1024 ~ 512 Solved Problem 1 104 A room has three lamp sockets. From a collection of 10 light bulbs, only 6 are Eood. ‘A person selects 3 at random and puts them in the sockets, Whar is the Probability that room will have light?1.124 Probability and Random Processes Solution Let X be the num¥eraf goo! bulbs out of 3. p = P(good batts - 74 16 z 4 (ey) > 41-55 P(x =2)= °C \70) \10 east one bulb is good Let B = Event that at I P(X <<) P(B) = P(X 21) =3- =1-P(X =9) 3 ae . [ a G 10 64 1000. = 10.064 = 0.936 1 Solved Problem 1.105 ‘An irregular faced dice is five even numbers is twice How many times in 10,000 even numbers? e probability that in 10 throws it will give lity that it will give four even numbers, ould you expect it to give no thrown and th the probabi sets of 10 throws each W‘ Solution Let X be the random variable representing the number of even numbers got in 10 throws. Given n=10 P(getting x even numbers) = P(X = 2) ? = 0G, prqlo-* Given P(X = 5) = 2P(X =4) 0G 5p%q° = 2 (°Cap'a®) ce 252p%q° = 2(210.p1q®) eae = 3527 pt+q=1 420 dea? tg=1Random Variables. 1.125 >. q=0.375 Hence p = 0.625 P(getting x even numbers) = P(X = x) = 19°C, (0.625)7 (0.375) 10-7 49 9(0.625)°(0.375)!°-° (0.375)29 . P(X =0) The number of times we get no even number in 10,000 sets = 10000 x (0.375)!° = 0.54993 = 1 approx Solved Problem 1.106 A department in a workshop has 10 machines. which may need adjustment from time to time during the day. Three of these machines are old, each having a probability of = of needing adjustment during the day and 7 are new, having ee 1 corresponding probabilities of —. ‘Assuming that the machine needs adjustments on the sarne day, determine the probability that on a particular day. 1. Just 2 old and no new machines need adjustment. 2. Just 2 machines need adjustment which are of the same type. Solution \ pi = (Probability of old machine needing adjustment) = + 10 > a=l-m=i P2 = (Probability of new machine needing adjustment) = x = 2-2 21 pi(X =r) = Probability of ‘r’ old machines needing adjustment 0" P(X =1) = Probability of ‘r’ new machine needing adjustment 7 Nn 20 a (x) Gr1.126 Probability and Random Processes P (2 old and no new machine needing adjustment) = pi(2)p2(0) ° a3 10 to, (2) (22)’ 2) (= a) ey" Color) \ar = 0.016 2. P(2 machine feeding adjustment of same type) = P(2 old or 2 new) = ni) _p2(0) + pi (0)-P2(2) oO =" (4) (#)” x ‘Cy (x) Gr a) + oo(zy 7) (3) x702(4)" yy = 0.016 + 0.028 = 0.044. Solved Problem 1.107 Aand B playa game in which _ A’s chance of winning at least t their chance of winning are in the ratio 3 : 2. Fing hree games out of five games played. Solution ~ P[A winning the game] = P(B winning the game] = a ci eet Loo n Let X be the number of games A wins. The probability of A winning r games is mveanre() QO)” PIA wins at least 3 games out of 5] = P(X > 3) = P(X =3)+ P(X =4)+ P(X =5) QQ “(J QrRandom Variables 1.127 Solved Problem 1.108 Assuming half of the population are consuming rice, so that the chance Of an individual being a consumer is } and assuming that 100 investigators take each 10 individuals to check whether they are consumers. How many investigators do we expect to report three or less were consumers. - Solution Let _X be the random variable representing the number of consumers of rice in 10 individuals . P(X=2)= 0G, (3) (Q) Sf yoty (2) - 2. (2) P(X <3) = P(X =0)+ P(X =1)+ P(X =2) + P(X =3) i 1 10 10; 1 10 H 1 10 fo 1 10 7 1 1 1 Co (3) + Oy (3) + PQ, (3) +e, () a 10 G (14104454120) 2 i@ - (3) (176) = ats = 0.1718 Among 100 investigators, the number of investigators who report that 3 or less were consumers = 100: x 0.1718 = 17 Investigators. Solved Problem 1.109 (AU, May/June 2006, MA040) A communication system consists of n components each of which will indepen- dently function with probability p. The total system will be able to operate effec- tively if at least one- half of its components function. For what values of p, is a 5 = component system more likely to operate effectively than a 3 component system. Solution a Since p is a coristant and the n components function independently, the number of components X that function follows a binomial distribution. P(z) ="Crp*q"*, x=0,1,2,...01.128 Probability and Random Processes P (5 component system functioning effectively) = PIX = 3ordor5] = P(X = 3) + P(X = 4) + P(X =5) = C5 p) gh + CaP Po" 4 OCs ph ght =1pgtrsrarr P (3 component system functioning effectively) = P(X = 20r3) = P(X = 2) + P(X =3) | = 30 p? gh? +3Cs Pa =3p*qt+p* lp @e+spiatr23r abr (Opa? + 5p*q + p* — 3q~p) 20 PP(LOp(1 — )? + 5p*(1 — 9) + P* ~ 31 —p) —P) 20 p?(6p? — 15p? + 12p — 3) 2 0 => — 3p*(p—1)2(2p—1) 20 st ee 3 Solved Problem 1.110 (AU, Nov / Dec 2007, MA 1253) It has been claimed that in 60% of all solar heat installations, the utility bill is reduced by at least one-third. - Accordingly, what are the probabilities that the utility bill will be reduced by at least one - third in at least four of five installations. Solution Here success is the reduction in the utility bill, by at least one third. Let X be the number of installations in which the utility bill is reduced by at least one - third. ~ X follows binomial distribution with ; n=5, p=0.6, q=0.4 P(X = 2) = °C,(0.6)*(0.4)°-*, « = 0,1,2,3,4,5 P(X 2 4) = P(X =4,5) = 5C4(0.6)4(0.4) + 5Cs5(0.6)9(0.4)° = 0.3370 Solved Problem 1.111 The Probability of a man hitting a target is 1. How many times must he fire $° Probability of hitting the target at least once is more than 90%. (AU, Nov/Dec 2003 MA039)Random Variables 1.129 foot the number of times he hits the target oe = => Then X follows a Binomial distribution, with P(X =2) = "Cer gh ="C, (3) G)* Given P(X 2 1)>0.9 » qofindn 1-P(X <1)>09 —P(X <1)>-0.1 v P(X <1)<01 2. P(X =0).< 01 “a Qn , (Jen n log ge Jog 0.1 n> 5.65 n>6 -:, He must fire at least 6 times. Solved Problem 1.112 (AU, May/June 2006, MA1252) lfm’ things are distributed among ‘a’ men and ‘b’ women, find the probability that the number of things received by men is odd. Solution 9 = Probability that a thing is received by a man = = q=l-p= 4 is the probability that a thing is received by a woman. The probability that out of ‘m’ things exactly x are received by men and the ‘itis received by women is given by p(t) =™C, p? g™—*, x =0,1,2,...m ‘hired probability = p = p(1) + p(3) + p(5) +--+ =™C pq?! + C3 pg? +s peg” > +---()1,130 Probability and Random Processes Now gh mesg? pt "Ce Tp mo, qh) p+ C2 q"? pe ee m1 4 MO pP g™-3 4 ay" (q+p)" = (q—p)" =" (qa +p) — (q— Py” = 2 (MO PF Butg-+p=landg—P = pee “+, By substituting in (1) 1 (b-a)™ -(Sae peg |t- prayer PLS Bera Solved Problem 1.113 Ina binomial distribution consisting of 5 independent tcials, probabilities ively. Find P(X = 3). 2 success are 0.4096 and 0.2048 respect Solution ‘ P(X =2) = Cop? 7,2 = 0,4,-°°5 given P(X =1)= 5Cypig4 = 0.4096 P(X = 2) = °Cop’q? = 0.2048 5pq* = 0.4096 10pq? = 0.2048 a7@s #£=2 2p i q=4p 1-p=4p : 1 5p=1, P=E 1\3 (4)? = 0.0512Random Variables 1.133 defective, determine the : 15, If 20% of the bolts produced by a machine are (@ 1, Gi) 0, Gii) atmost 2 probability that out of 4 bolts chosen at random, bolts are defective. (Ans: (i) 0.4096, (ii) 0.4096, (iii) 0.9728) 4.5 Poisson distribution 4.5.1 Definition If X is a discrete random variable that assumes the values 0, 1, 2, probability mass function is given by .. such that the 2 =0,1,2,3,..- P(X=2)= A>o 0: otherwise then X is said to follow the Poisson distribution with parameter . 4.5.2 Poisson distribution is the limiting case of binomial distribution . (AU, Nov / Dec 2008, MA 1254) Suppose in a binomial distribution, : 1. The number of trials is indefinitely large i.e, n — 0. 2, pbe the probability of success in each trial is very small i.e., p — 0. 3, np(= A) is finite and p = a. q=1-p= Now for binomial distribution P(X =2)=ne, p> gq", 2=0,1,2,... _ n(n- (n= 2)... (n-2 4 a a! = MOD (n= 2). 240 OF 0-2)" - [@-2) (@-2)...0-23 0-27 6-3) ]1134 Probability aid Random Processes: Tuking limit on both sides. sting (a) = [ris (( 1) (: for 20, 1,2,.06 Which is the p.m.f of the poisson distribution mayors ae 8)" n00 1.5.3 Moments about origin of Poisson distribution pi = BIX] OF My = E[X) = ai The mean of the Poisson distribution is .\ Now B1X4] = > 22 p(x)Random Variables 1.135 Eee S: ete py = E{X?| Variance (X) = E(X?) — (E(X))? = “i ? — (A)? Var (X) = E[X3] = e* Ee Y ele —1(«@-2) +32(x-1) +4] 70 =e [+3 +} ed =N43N TA ant =e Siete ~ Ile —2)(e— 3) + 6x(@ — Ye — 2), Z=0 “4122 -1) +a% =e MAGGEP ROB ETEM VOX +AC* pi, = BIX4] = 4469477 +2 15.4 Moment generating functidn (M.G.F), Mean and Variance (AU, May/June 2006, MA 040) (AU, Nov/Dec 2008, MA 1253) Mx(t) = E[e*] = > e* plz) a ayes z=0 eo or1,136 Probability and Random Processes . My) set [seven], = [ean dt t=0 = ete =A Mean = t-1)y ot . ferea ety(rct) +e re | to eu, SEE _Me-1(Q €0)(A €0) + ENE -D(A C9) E[X7]=N +A Variance (X) = B(X?] ~ (21%) Atv? = (A)? +, Var (X) =A Note: , : Examples of random variables that usually obey the Poisson distribution, 1. The number of misprints on a page. (or a group of pages) of a book. 2. The number of people in a’community living up to 100 years of age. 3. The number of wrong calls among telephone numbers that are dialled day. 4. The number of biscuit packets sold in a particular store each day. 5. The number of customers entering a post office on a given day. 6. The number of vacancies occurring during a year in a particular depattn 7. The number of a— particles discharged in a fixed period of time ft radio active material. ‘8. The number of cars Passing - aday. 9. The number of suicides Teported by a particular city. ‘10. The number of accidents in some unit of time. accrossing per minute during the busy ho!Random Variables 1.137 Note: ‘The Poisson probability distribution provides good approximation to binomial probabilities when n is Jnrge and \ = np is small, preferably when np < 5. Recurrence relation for the moments of the Poisson distribution a a” iF P(X =2) = 2=0,1,2,... 1.5.5 By definition of moments , Mr = E[X — E(X)\ =E[X—-Ad" [." E(X) = Mean = d] = @-d)" v2) m= = e~A \E br = So (e@-d mane z=0 tA Differentiate the above w.rt A, we _ en Filed = Da a Ee ») | = So 5 [@- d)F4(=1) e7> At +(z- AJ eA(-1) AF + (2A eA a] “= A)T~1 ew nF eo rete se wed (2 —r(z— Ay} o> dF + (a d)Fe 07 G- 1)] W Bye = BIs - *) ere ayteaae] (enn _, Senate é al i[e = d)re nz ( ( i} YI >i yf a ; j TT Mra1.138 Probability and Random Processes +r pra = yee d jtrt = [ax (pr) + rt] d at on distribution (AU, May/June 2006, MAt254) ; i ith parameters If X and Y are independent Poisson variates My Pi tek pen Ag Respee, tively, then X + Y is also a Poisson variate with parame 1 2+ 1.5.6 Additive property of Poiss Proof X is a Poisson variate with parameter Ai _ Mx(t) =e ¥ is a Poisson variate with parameter Az My(t) = ee“) Now Mxsy(t) = Mx(t) - My(#) = erlet-D . er2(e'-D [. X and ¥ are independent] = er tr2)(ef-1) +. X4Y isalso a Poisson variate with parameter A; + 2. Solved Problem 1.114 ‘ (AU, May/June 2006, MA1254) If X1 and Xo, are independent Poisson variates, show that X, — Xo is nota Poisson variate. Solution Let X and X2 be independent Poisson variates, having parameters \, and 42 respectively. Now Mx,~x2(t) = Mx,(t) - M_x,)(t) = Mx,(t)- Mx,(—t) By property of mgf = edlet=2) , pdalem!-1) The above cannot be expressed in the form e*(e'+)) ©. Xy — X2 is not a Poisson variate. .a Random Variables 1.139 solved Problem 1.115 If X is a Poisson random variable such that P(X = 1) = 7 P(x = 2) =4 find P(X = 0) and P(X = 3). solution Ayr PX #2) a " a 3 Given P(X =1)=5 (X= D= 3G ALES > = 1! 10 ee feo er 10y @ 1 P(X =2) == eS? 1 peer eae -ay2 _ 2 ; en = 5 (2) Substituting (1) in (2) ae G2 Tox* “5 20 4 A= Fp 5 a hss en3 P(X = 0) = _ 67139(1,33)8 ce 3! __ (0.264)(2.352) eee : : = 0.103 Solved Problem 1.116 AU, April/ May 2008, MA 1253 Let X be a random variablé following Poisson distribution such that + P(X =2)=9P(X =4) +90 P(X =6) Find the mean and standard deviation of X.1.140 Probability and Random Processes Since the random variable X follows Poisson distribution. X follows Poiss i id P(X =") = aT Given: P(X = 2) =9 P(X = 4) +90 P(X = 6) e8 a 90 r? 6! Cancelling e~*A? both sides 1. 92, 90A4 aa" 6 19 24 0 2° 24 720 1_3,2,1\4 =e =X 373 w+ 3 Mt 32 1 ete 9-0 M443. -4=0 A=1lor -4 A=1 [A is non-negative] Fora Poisson Distribution, Mean =Variance = Mean =X=1 Variance = \=1 Standard deviation = 1. Solved Problem 1.117 @ Neither cars were used (ii) Some demand is Tefused,Random Variables 1.141 solution Let X be random variable representing the number of demands for cars. AYE P(x demands ina day) =P(X =x) = Given \=15 : : Now © P(X Sn) a ant (i) The proportion of days on which neither car is used P(X =0)= eee, = e715 = 0.2231 (ii) The proportion of days on which some demand is refused. ‘The demand is refused when X is more than 2 P(X > 2) =1-P(X <2) 1- [P(X =0) + P(X =1)+ P(x = 2)) =1.5(1,5)0 e-15 st (— ye 2 G5) o! cc = -1.5 (1.5) =l-e [rvis+ GS" . 0.19126 Solved Problem 1.118 6 coins are tossed 6400 times. Using Poisson distribution, what is the approximate probability of getting 6 heads 10 times? Solution p = probability of getting six heads in one toss of six coins -( 1 6 d= np = 6400 x G) = 1001.142 Probability and Random Processes Let X be the number of times we get 6 heads e7100(100)'° P(X =10)=—4q,_ "= 1.025 x 10-° Solved Problem 1.119 The probability of an individual suffering a bad reaction from_an injection of certain antibiotic is 0.001. Out of 2000 individuals, find the probability thar a Exactly 3 suffer. (ii) More than 2 suffer from bad reaction. "@ Solution" Let X denote the number of individuals suffering from bad reaction p= 0.001; n = 2000 Since n is a large and probability of bad reaction is small we can assume X as a Poisson variate with \ = np = (2000)(0.001) = 2. ert 292 (@ P(exactly 3 suffer) = P(X = 3) 7 23 gla 0.1804 Gi) P(more than 2 suffer) = P(X >2) =1-P(X <2) =1-[P(X =0) + P(X =1)+ P(X =2)] e7229 e-291 252: ey etoen Ctra ema [ 7 He I Fg 2! | =1-e%f1+242] 1-5e? ( = 1- 0.6767 = 0.3233Random Variables 1.143 Solved Problem 1.120 i ‘A manufacturer produces IC chips out of which 1% are defective. Find the probability that in a box containing 100 chips with no defectives are found. Solution Here =i : 0.01 Since n is large and p small, we can use Poisson distribution A yz P(X defectives) = P(X = 2) = £—* Here, Nesp! oot : e110 P(X =0)= = e7) = 0.3679 Solved Problem 1.121 A machine consists of 2000 equally reliable parts with a probability of failure for cach of them equal to 0.0005. What is the probability that the machine will fail to operate if failure occurs when, at least, one part fails to operate? Solution Let X be a random variable of number of failures. Total number of parts n = 2000. Probability of failure p = 0.0005 Hence A=np= 2000 x 0.0005 = A yr P(X=z)= = — P(X >1)=1-P(X <1) =1-P(X =0) 140 = =1-e71 =0.6321 Solved Problem 1.122 (AU, May/June 2006 MA 034) AU, Nov / Dec 2008, MA 1252 In a company the monthly break down of a machine is a random variable with Poisson distribistion, with an average 1.8. Find the probability that the machine will function for a month (i) Without break down, (ii) With exactly one break down, (iii) With at least one break down.1.144 Probability and Random Processes Solution Let V denotes the number of break downs in a mon! th. Now PIX Given mean P(X =) Tg (i) P (no break down) or! 8(1.8)" _ 18 = 0.1053 = P(X =0)= aI | | (ii) P (exactly one breakdown) i -1.8(1,8)! . =P(X=1)= “or = 0.2975 (iii) P (At least one break down) = P(X >1)=1-P(X <1) — P(X =0) = 1— 0.1653 = 0.8347 Solved Problem 1.123 ‘A book of 500 pages contains 500 mistakes. Find the probability that there are a east four mistakes in a randomly selected page. Solution “Total number of mistakes = 500. Total number of pages = 500. There is an average of 1 mistake per page i.e., A = 1. Let X be a random variable of number of mistakes in a page: A yz -1y42 P(X =2) = al P (at least four mistakes) = P(X 2 4)=1- P(X <4) =1-P(X <3) =1- (P(X = 0) + P(X = 1) + P(X =2) + P(X = 3)Random Variables 1.145 “3 [ee Soy , or, “ar = 11 414342 e [2+ +36} = 0.0189 Solved Problem 1.124 - A life insurance company insures the lives of 5000 men of age 42 years. The probability of a 42 year old man dying in a given year is 0.001. What is the Frobablity thatthe company will have to pay 4 claims in a given year? Solution Here the Total no of men insured = n = 5000 : Probability = p = 0.001 A=np=5000 x 0.001 =5 Since ‘n! is large and probability is small, we can use Poisson distribution Cyakes | emits a al P(X =2)= 554 P(X =4) = £—* = 0.1755 A! Solved Problem 1.125 ‘A manufacturer of wet grinders wants to buy-one-hp motors from a supplier, in alot of 1000. When fitted to the machine, these motors have the probability of failure 0.001. In a shipment of 1000 motors what is the probability that (i) None are defective. (ii) One is defective, (iii) Two are defective, (iv) Three are defective. Solution ~ Aye P(X =2)= n= 1000; p = 0.001; .P(X=2)= oS" P(X =0) =1.146 Probability and Random Processes = 0.184 = 0.061 Solved Problem 1.126 (AU, May/June 2009, y4,. The average number of traffic accidents on a certain section of a highwa,. per week. Assume that the number of accidents follows a Poisson distr, Find the probability of (i) no accident in a week (ii) atmost two acciden, week period. i Solution The average number of ‘accidents per week is A= 2: The probability of x accidents during a week et =P(X=2)="5 @ P (no accidents) = P(X = 0) oF 42% = = ee a * = 0.1353 (ii) During a2 week period the average number of accidents on this sect highway would be 2 + 2 = 4, by additive property of Poisson distrib: The probability of atmost 3 accidents during 2 - week period. P(X <3) = P(X =0)+ P(X =1) + P(X =2) + P(X=! e-4 4o e4 4l e-4 42 —4 3 Caren Chuan e 44 ot ta ta met[i44 16 64 : +3+G = e~4(23.67) = 0.4335Random Variables 1.147 oived Problem 1127 . : sata Poisson distribution to the following data and calculate the expected frequen cies: Deaths 0 1 2 3 4 Total Frequency 122 60 15 2 1 200 Eq where N= of = 200 By equating the mean of the given frequency distribution to the mean of Pois- son distribution, the expected frequencies are obtained as follows. : e eA d= x “f P()=T= Expected frequencies [N.P(x)] 0 122 0.60653 121.30 = 121 1 60 . 0.30327 60.65 ~ 61 2 15 0.07581 15.16 = 15 3 2 ~~ 0.01263 2.53 =3 4401 0.00157. - 031~0 Total 200 Here we see that the expected and observed frequencies are equal and hence the Poisson fit is given by -95 . (0.5) P(X = 2) = P(z) =; 2 :0,1,2,3,4 Solved Problem 1.128 (AU, May/June 2006, MA 1252, MA 1254) If X and Y are independent Poisson random variables, show that the conditional distribution of X given X + Y is a binomial distribution. Solution _ | ‘Let X and Y be independent Poisson rar:dom variables, with parameters Ay and Xz respectively P(X =rNX+¥=s) P(X +Y¥ =s) P(X =rnY =s-7) ~~ P(K+¥ =8) Now = P(X =r/X+Y=s)=1.148 Probability and Random Processes 7 _ P(X =r)x PY =s-7) P(X +Y =8) (XE Y are independent) et Gay rt x Tea! Seeley = SSP Fa)! a son variables, with parameters Ai and dy, xe + 1 ees) Gy (Since X and Y are Poi a Poisson variable with parameter 11 P(X =r/X+Y =) => +A2) ql AL where p= cee which is'the pmf of a binomial distribution Solved Problem 1.129 Assume that the number of cars passing by a traffic junction obey a Poisson iy bution. If the probability of no cars in one minute is 0.20, what is the probaiy of more than one car in 2 minutes. Solution ? Let X be the r.v of the number of cars in one minute eat a! P(X =2)= en 9 P (of no cars) = P(X =,0) = oe a 0.2." (given) oop oe AL OQT ‘Taking log on both sides —A log e = log 0.2 af x= -k Aisa og (3) log 5 = 1.6094 For one minute \ = 1.6094. For two minutes A = 2(1.6094) = 3.2188Random Varlables — $.149 er” be the FY of number of ears in 2 minutes then i PY >1)=1-P(Y $1) =1-[P(¥ 50) + P(Y = 1) sig [= 1483.18) ’ aegis | oO! w = 1— e757188 (1 + 3.2188] = 1—0.1687 - = 0.8312 Solved Problem 1.130 ‘an insurance company found that only 0.01% of the population is involved in road accident in a particular area. If 1000 policy holders were randomly selected the population, what is the probability that not more than two of its clients from : will be involved in such an accident next year. Solution Given p=0.01% 0.01 e100 0.0001 n= 1000" A= np = 0.0001 x 1000 = 0.1 Let X be ar.v of number of clients involved in accident edt 0.19.1)? a! al P(X =o) P (not more than two) = P(X < 2) P(X =0)+ P(X =1) + P(X = 2) .1(9,1)° - e781(0.1)! 7910.1)?1.150 Probability and Random Processes Solved Problem 1.131 Suppose that the number of telephone calls coming ig 2 enone eg between 10 a.m and 11 am, say X, is a Poisson vane! pee 1=2, Sing! calls arriving between 11 a.m and 12 noon, say Y,, is a Poisson variate wig 6. If X and Y are independent, what is the probability that more than 5 cals between 10 a.m to 12 noon. Solution Given: X ~ P(2) Y ~ P(6) By additive property Z = X + Y ~ P(2+8) 3 Z=X+Y~P(8) er . e887 ar z=0,1,2,... . Hence a ~~ 2 P (more than 5 calls between 10 a.m to 12 noon) =P(Z>5)=1-P(Z<5) 5 8 gs oe zl 2=0 — 0.1912 = 0.8088 Solved Problem 1.132 Let X be a random variable which assumes Poisson distribution. @ Find P(X = 4) if P(X = 1) = P(X =2) (ii) Find E[X] if 2P(X = 0) + P(X = 2) = 2P[X =1] (ii) If \ = 1 find P(X > 2/X <4) Solution P(X =2)= ° When P(X = 1) = P(X = 2) : eA eA 2 ee oe oeRandom Variables 1.151 (i) OF 9 p[X = 0] + PIX =2)=2P[x = 1] A ena At eA I oto ae 2 > o (249) = 20a => A+ =4r => N-4\44=0 => AS EX] =2 -142 Gi) A=1> P(X =2)=} | “4, -P@sx <4) To find eS 28 areca : a P(X =2) + P(X = 3) + PK =4) ye. te Solved Problem 4.133 ‘A manufacturer of cotter pins knows that 5% of his product is defective. If he sells cotter pins in boxes of 100 and guarantees that not more than 10 pins will be defective, what is the approximate probability that a box will fail to meet the | guaranteed quality. Solution Given n= 100. = Let p = probability of defective pin = 5% = 0.05. Hence A = np = 100 x 0.05 = 5. : Since n is large and p is small we can use Poisson distribution. Now P (x defective pins in a box of 100) =P(Xa2)=2-™, 2 =0,1,2. al P(X > 10) =1—P(X < 10)1,152 Probability and Random Processes 10. 4-8 5 al 2=0 55> - oo =1-e yan - Solved Problem 1.134 A machine produces a certain item with a defectiveness of 1%. By PPlying son approximation show that the probability that a sample of 100 items Se at random from the total output will contain not more than one defective jg 2/e. Solution Here p=0.01; n= 100 => A=npH=1 P (x defective items) P (not more than one defective) = P(X < 1) = P(X =0)+ P(X =1) le nl shy eaT trance tres Solved Problem 1.135 AU, May/June 2009, MA 12! Ina book of 520 pages, 390 typographical errors occur. Assuming X is a Pois« variate for number of errors per Page, find the probability that a random samp of 5 pages will contain no error. Solution ‘The average number of typographical errors per page 3s 390 _ O75 0.75 (9,75) alRandom Variables 1.153 P (1 page contains no error) = [P(X =0)] = [ = [er2*5] -. PIS pages contain no error] = (e~%75)5 — e-3.75 — 9.9235 solved Problem 1.136 After correcting the proof up to SO pages of a book, the proof reader found that there are three errors per 5 pages. Use Poisson distribution and estimate the num- per of pages with 0, 1, 2, 3 errors and more than 3 errors in a book of 1000 pages Solution A = average no of errors =0.6 5 Let X be the r.v of number of errors in a page. Aa eX 096.6)" fxs zl No. of pages having no errors in that book ~6(0,6)° = 1000 x P(X = 0) = 1000 [=] = 1000 x 0.5488 = 549 pages + No of pages having one error = 1000 x P[X = 1] = 1000 [ = 1000 e~°S x 0.6 = 329 pages No of pages having two errors “ Te-9-6(0.6)? = 1000 x P(X = 2) = 1000 ai = 99 pages1.154 Probability and Random Processes . No of page having 3 errors : 0. (0.6)5 = 100 x P(X = 3) = 1000 (Gr = 20 No of pages having more than 3 errors = 1000 — (549 + 329 + 99 + 20) = 1000 — 997 = 3 pages Solved Problem 1.137 In a certain coal mine the probability of a worker dying in an accident in a given year is a Let X be a random variable of number of workers dying in an accident! which assumes Poisson distribution. If there are 350 workers in that mine find the probability that there will be at least one fatal accident in a year, a Solution P(X =2)= < — Given - . P= Tapp) = 350; 1 = Tapp * 350 = 0.25 e79?5(0,25)* aa ee “. P(X =0) = 79% = 0.7788. P[X > 1)=1-P[X <1] =1-P[x=9 ee = 1-795 — 0.2212 Solved Problem 1.138 (AU, May/June 2006, MA 1253) Inacomponent ‘manufacturing industry, there is a small probability of ae for any Capa 2 hoideeetive. The components are supplied in packets of 10. Us¢ on to calculate the approximate number of packets containing (A) no defective, (B) one defective i i eigen cine component (C) two defective components inRandom Variables 1.155 solution “Gives P= 500° Let X be the number of defectives in a packet n=10. ae e7°92(9.92)= ee The number of packets free from defect = 0.9802 x 10000 = 9802. (B) One defective component: 0.02, 1 P(X =1)= “Oy = 0.0196 No. of packets with one defective = 0.0196 x 10000 = 196. (©) Two defective components: E 0.02, 0. P(X = 2) = (0-02? —_ 9 coo196 No. of packets in which two components may be defective = 0.000196 x 10,000 = 1.96 = 2, Solved Problem 1.139 i (AU, Nov/Dec 2002, MA040) Ifa Poisson variate X is such that P(X = 1) = 2P(X = 2). Find P(X = 0) ‘and var(X)., Solution Given a Poisson variate, Let’ be the parameter1.156 Probability and Random Processes Given P(X = 1) = 2 P(X = 2) = a? -A=0 MA-1)=9 y= Since A>O | To find { AA eh 0.3679 | P(X =0)= i i Var (X) = A= | | Solved Problem 1.140 (AU, April/May 2004, MA g4y Jt is known that the probability that an item produced by a certain machine will be defective is 0.05. If the produced item was sent to the market in packets of 20, find the number of packets containing at least, exactly and atmost 2 defective items in a consignment of 1000 packets using Poisson approximation. outer Giveri n= 20; p= 0.05 A=np=l X is a Poisson variate eo ry P(X =2)= xe a eal ! @) P(X =2)= =z = 0.1839 .. Number oft packets containing exactly.2 defectives = 1000 x 0.1839 = 184 (nearly) POR 2 2) = 1- [P(X = 0) + PUK = 1)] = 1- = 0.2642Random Variables 1.157 Number of packets containing atleast 2 defectives C 1000 x 0.2642 = 264 P(X = 0) + P(X =1) + P(x =2) el Gi) P(X $2) +, Number of packets containing atmost 2 defectives = 1000 x 0.9197 = 920 nearly —_—_— Exercise 1.2 1, A wireless set is manufactured with 25 soldered joints each. On an average one joint in 500 is defective. How many sets can be expected to be free from defective joints in a consignment of 10,000 sets. (Ans: 9512) 2. Red blood deficiency may be determined by examining a specimen of the blood under the microscope. Suppose a certain small fixed volume contains ~ onan average 20 red cells for a normal Person. Using poisson distribution, obtain the probability that a specimen from a normal person will contain Jess than 15 red cells. > 779(20)= ==0 a 3._A Hospital switch board receives an avera ge of 4 emergency calls in a 10 minute interval. What is tho probability that (i) There are at most 2 emergency calls in 10 minute interval. (ii) There are exactly 3 emergency calls in 10 minute interval. s 32) Ans: (i) 13e~4, (ii) ($) | 4. In company producing optical lenses there is a small chance of 1/500 for any lenses to be defective. The lenses are supplied in a packet of 10. Use Poisson distribution to calculate the appropriate number of packets contain- ing ( no defect, Gii) one defective lense, (iii) two defective Tenses, (iv) three defecti in a consignment of 20,000 packets. [Ans: (i) 19604, (ii) 392, (iii) 4, Gv) 0]