Scribd
Upload
14 views19 pages

Download

dow

Uploaded by

Ruby Smith
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
14 views19 pages

Download

dow

Uploaded by

Ruby Smith
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
Chapter 16 fm is latge, the evaluation ofthe binomial probabilities can ievolve considerable computations, In such a case a simple apptoximation to the binomial probabilities could be of considerable use, The approximation P is close to zero is called the Poisson. inomiabwhen n is large and p is not distribution which willbe discussed Hered = np Note\}+ The binsmsial distelbution is characterised by two parameters 2 1 stile the Poisson distribution is characterised by a single parameter A. ‘The Poisson distribution was first discovered by a French mathematician S.D. Poisson in 1837. ‘The sample space for the binomial distribution is {0, 1,2, ..n} ‘hile the sample space for the Poisson distribution s (0, 1,2... Poon} Note 4: Expected value (Mean) is given by A and variance of the Poisson distribution is also 4 Additive property of Poisson random variables IE, and X, are two independent Poisson random variable with | Parameters A, and 2, then X, + X, is a Poisson random variable with ParameterA, +4, Poisson frequency distribution ‘Let a Poisson experimental consist ofn independent trials, Let this experiment, under similar conditions be repeated N times. Then Poission Distribution 85 nay : eae iber of x successes in N experiments, Mee 1" gives the expected numt a i ther with isting ofn trials, The possible number of successes toget diate peer rcnouce oat 6 cabiuamirtmes reat} distribution. The following arc some of the examples where the Poisson probability lawcan be applied : (1). Number of defective items produced in the factory. (2) Number of deaths due to arare disease. G)_ Number of deaths due to the kick of a horse is an army. (4) Number of mistakes committed by a typist per page. variable such that Example 1: X is a Poisson random P(X=1) = 03 and P(X=2) = 02 Find P(X -=0). Solution: 1fX isa Poisson vandom variable with parameter 1 Dividing (2) by (1) we get, A =4/3, » PUK=0) = et = eo = 0264 it tive Fuses will Example 2 : Find the probability that atmost 5 defective fuses wil be found inne of 20 usa ‘ifexpericnce shows that 2 percent of such. fuses are defective, given that e~* = 0.0183, Solution: P= Probability that a fuse is defective A=np=4 Probability that atmost 5 defective fuses willbe found in a box of 200" fuses, = PKs) WWW .[Link] Business Staisties iis i w 643) anes (2) 078s Example 3: The number of accidents in a year attributed to taxi rivers in a city follows a Poisson distribution with mean equal to 3. Out ‘of 1,000 taxi drivers find approximately the number of drivers with () —noaccidents in a year Gi) more than 3 accidents in a year. Solution : Number of drivers with no accident = 1000 x 0.05 = 50 Oe heen on fe a ciate ee eer ee ora Scie erent eee random variable with parameter 4, If these two random variables are independent what is the probability that more than 5 planes cross the border between 5 pm. and p.m. ? Solution: Let_X, = number of planes crossing the border between 5 p.m, and 6 p.m, Porssion Distribution 8) "and X, = number of planes crossing the hordes between 6 p.m and 7 p.m, Since X, and X, are independent Poisson random variables with parameters 3 and 4 respectively, X, + X, is a Poisson random variable with parameter 7. P(X, 4+X,>5) 1 —033007 = 0.46093 Example § : From records of 10 Indian Army corps kept over 20 years the following data weré obtained showing the number of deaths caused by the horse, Calculate the theoretical Poisson frequencies No. of deaths Gey es ay tet Frequency ieee eee a te Zht Solution : = 2 OFS HM FOG 3 200 061 The theoretical frequencies are given by, 200.e* A whore x = 0, 12,3, a 41,23, and 4 = 061 No. of deaths ON 273i) ba oreuak observed frequency 10965) 22-31 0 Theoretical frequency: 109 - 66 2 —4 1 2% Example 6: 1f3% of the electric bulbs manufactured by accompany are defective, find the probability that in a sample of 100 bulbs exactly five bulbs are defective, “3 = 0498) WWW .[Link] Pp ~ Probability hat a bulb is defective ~ 08 = 1000 xe aan ‘ : LN, (i) No. of pages with two errors z ees 1000 x p @) 1000 x e~$ = 1000p (3) 2 = 1000 x «8 = 20 pages Example §: In turning out certain toys in a manufacturing process in a factory, the average number of defectives is 10%, What is the Example 7 : After correcting the proofs of the first 50 pages «1 probability of getting exactly three defective toys ina sample of 10 toys ‘book, iis found that on the average there are 3 errors per 5 pages Poicsion probabilities and estimate the mumber of pages with 0, | 7 ‘chosen at random by using the Poisson distribution. (e~' = 0.36788) errors in the whole book of 1000 pages, ¢~"* = 5488) Solution p= Probability hat atoyis defective Solution : A= Average mumber of errors per page = 0.0, n = 10 3g A= np = x(a) = 1 +. The probability that there ax errors per page is pe) =€ Probability that there are 3 defectives arts ego) ee eas Sad Py Example 9: A manufacturer of pins knows that 2% of bis products are defective. Ifhe sell pins inboxes of 100 and guarantees that not more than 4 pis will be defective, what isthe probability tha a box wil filto ([Link] pages containing no error meet the guaranteed quality ? (e~? = 0.13534) = 1000p(0) = 1000 6" 6 = 1000 > 5488 Solution: p= Probability that apinis detective = 325 = ou? = 548.9 = 549 pages n = 100 (i)__No. of pages with one error A= np = 100.02 =2 = 1000 x p(t) WWW .[Link] Business Statistics ‘The probability that a box chosen at random will fail to meet the guaranteed quality para 1-p&s4) 1 [p() + pa) + PQ) + pO) + pC] | ious [ie aad 413534. 7 1052682 Example 10; An insurance company has discovered that only about (0.1 percent ofthe population is involved in a certain type of accident each year. If its 10000 policy holders were randomly selected from the population, what isthe probability that not more than 5 ofits clients are involved in sueh an aecident next year ?(e~! = 000045) Solution: p= Probability that a person will involve in an ca accident ina year = 555 = 799 = 10,000 sp = spn sds 30 Probability that not more than 5 will involve in such an accident in ayear pass) [r@) + pay + rey + rd) + o(4) + POD] Poissivn Distribution — =10) 10 10 10° eli 24a ae [tee = oonss [i + esa 4 5 ey 2H ems (49) = 161, Example 11 : Suppose the probability that an item produced by a particular machine is defective equals 0.2. If 10 items produced by this machine are elected atrandom what is the probability that not more than ‘one defective item is found? Use the binomial and Poisson distributions. Solution : a = Probability that an item proddced is defective = 0.2 4 = 08, 10 The probability that x tems are defective is po) = 10¢, (2 (8) Pobabilty that not more than one is defective (0) + pc.) = (8) + 10C, (2)(8? = 3758 tox2=2 Pp) Probability that not more than one is defective pO) +p) =e? WHA 0602, Example 12; The probability that aman aged 35 years wil die before teaching the age of 40 years may be taken 280.018, Out of a group of 400 rmen now aged 35 years, what is the probability that 2men will die within the next 5 years ?(e~? = 000912, e? = 8187) Solution: p= 018 n = 400 WWW .[Link] 40 Business Statsties A= np = 40x 018 = 72 = 72, 2 py = 2 OF Probability that 2 will die within the next 5 years = e772 (12 ?Q “F =x 2592 = 0.01933 Example 13: A factory employing a large number of workers finds that over a period of time, the average absentecs fate is three workers per shift, Calculate the probability that in a given shift (9) cxatly t¥0 workers will be absent (more than four workers will be absent (e' Solution: = 2 = 3, 04979) 2G) = 6, x = 01,2. cas W) Probability that exactly 2 will be absent Pere a aes (ii) Probability that more than 4 workers will be absent = P&>#) = 1p (Ks 4) += 1 [PO + PA) + pe + ) + pA] ee 4 es [Lea 44s +454 a3] = 1-099 « 16375 = 1-853 = 1847 Example 14 :Ifa random variable X follows a Poisson distributi such that P[X =2] = P[X = 1]. Find P{X =0] Poission Distribution Sotution Given p2) = p(1) 1B pO) =e? = 13534 Example "5 : /ssume that the chance of an individual coal miner i eid ing a year is <1 oissont being killed i mine accident during a year is 7755. Use the Poisso distribution ,o calculate the probability that io a mine employing 350 miners, he-e will be atleast onc # tal accident in a year. 775 = .7788) Solution Pw = Bd pan = = pe = The probability that there will be atleast one fatal accident = P(X 2 1) P(0) eo = 1-788 = 0212 Example 16: At a besy traffic junction the probability p of an individual having on accident isp = .0"1, How ver during a certain part ff the da. 1000 cars pass through the junction. What is the probability that owe or more accidents occur during that period ? (e~? = .9048) Soluti mm P= 001, 1 WWW .[Link] Business Statisties Poission Distributio probability hat 2 or mre accidents in a year (Gi) Number of packets containing two defectives = pee? Np) i = 1= [pa A 2 (02 1 [pe + pay) = 1,00,000 x e rs 2» Shot x AL = 0047 EXERCISE 16.1 Example 17: One-fith percent of the blades produced by a blade manufacturing factory turn out to be defective, The blades are supplied Ifxis a Poisson variable such that P(X =2) = 9 x P’=4) + 90, in packets of 10, Use Poisson distribution to calculate the approximate P(X'=6), find (j) the mean 2 of X (i) the mode of X, (i) the funber of packets containing an defective, one defective and wo coefficient of skewness lefective blades respectively in a consigoment of jackets, Gite | ee a Tna certain factory turaing out razor blades there is asmall chance ‘, ‘of 0.0472 for any blade to be defective, The blades are supplied in packets of 10, Use the Poisson distribution to calculate the approximate number of packets containing no defectives, one fective and two defcetives respectively in a consignment of 10,000 packets. Suppose the probability that an item produced by a particular machine is defective equals 0.2. If 10 items produced from this machine are selected at random, what is the probability that ot ‘more than one defective is found? Use the binomial and Poisson (0) Nuraber of packets containing no defective istributions and compare the answers. Solution : = Np(o) ‘Am office switch board receives telephone calls at the rate of3 calls eh per minute on the average. What is the probability of receiving = 100,000 6) callsin one minute interval, (ji) atthe most 3 cal ina minute 1,00,000 x 9802 interval? = 98020, (ii) ‘Number of packets containing one defective =.Np(1) = 92 02) 1,00}000 «27% WWW .[Link] iz Business Statisties ‘Obtai the mean and variance of poisson distribution Solution? sh fate pte] 2 fay variance of x= E Baha 2 Ina poisson distbution mean = variance = Simta!y We can show that the jij =m and yg = 3 ¢m Exercise 16.2 Records show thatthe probability is 0,000102 that a car will have» i type while driven over this bridge, what isthe probability that not than off Will have a flat type? Dens aa WWW .[Link] is, ees te pti atina fy engine 20 vas there will be atleast two fatal accidents in a year. Given (e 7 = 0 7788) ‘A local electrical appliance shop has found from experience thatthe demand for tube lights is distributed as poisson with a mean of 4 tube lights per rweck, If the shop Keeps 6 tubes during a particular week, what is the probability that the demand will exceed the supply during that week? (Given e- =0.0183 12 percent of electric bulbs manufactured by the certain company are defective, find the probability that in a sample of 200 bulbs. (less than 2 bulbs, (i) more than 3 bulbs are defective Given ¢* =0,0183) [Number of road accidents during a month follows a poisson distribution ‘with mean 6, Find the probability that in 2 certain merth numberof acidemts will be (@ Not more shan 3, Given e* = (0.0024) {ffora poisson variance X, P(X =1)=P(X=2) find P(¥=1 01 2). Also find its mean and variance. ‘A discrete random variable X follows poisson law. Find P(Y >2) and PV is at most 2), if itis given that E(X)=2,5 and ¢"** = 0.0821 U£58%of the families in Calcutta use gas as fuel what wil be the probability of 10 families using gas as ful ina sample of LOD families selected at random among the families in Catcutta?. You may assume poisson distribution? Given (¢~ 0 9067) (i) Between 2 and 4 Ina Company manufacturing foys itis found that 1 in $00 is defective Find the probability that there wll be at most two defectives in a sample of 2,000 units 10 A typist kept a record of mistakes made per day during 300 working days ofay Mistakes per day: [9 Roof days TRS Chapter 17 Normal Distribution The normal disribution which will be discussed now is the Most important probability distribution in statistics. It is = continyous distribution. 1 can alsobbe derived as the limit of the Binorsiai dist ‘under suitable statistical conditions namely (i) the sample size is °° Giebiverta tobe tat mterets) aes (ree aiairiay ett is all Several mathematicians were instrumental for the discovery of this distribution and among them the most prominent mathematician is Kart (Gaass and hence this distribution is called the Gaussian distributios The normal distribution occupies a prouninent place in Statéties use of its clegant properties and applicability (o many pravtieal siqations. Normal distribution plays a vital role in sampling teary for dro'ving conclusion about point estimate of parameters. The noimal distribution comes as abest approximation for fitting the actual obscsel frequency distribution of many phenomena, For-example, a noi distribution will be a close approximation to the distribution of heights or weights of individuals in @ homogencous population or to the distribution of tensile strength of pieces of steel produced by a ce/tal process. ‘Characteristic of Normal Distribution The diagram of a normal distribution is given below. tt i cil normal curve, Ic suggests several important properties of the nor 45 «eical distibution and the val distribution is b & shaped Phe curve has < single peak point de the distribution is unimodal. The mean of se normal distribu mn fies atthe centre of normal Becw se ob the symmeuy of die normal curve, the median and monte are abso atthe centre (th: normal curs: Hence in anormal isteibpion the mean, median and mode evincide Phe tails of the nocmrat distribution cxtend indefinitely and never ouch the horizontal aves. The" fs, we say thal the normal curve approche. asyimptorivally a enires side of its hor. one! axe (©) The normal dsiribstwn is a two parameter po sba istribotinn, The porameters mea art standard devati) completely leterinine the distyibusion. 7) Aces property Ins oormat distribution, hy a1 67% of the nbvervati Juiweon moun # SD (eye #9), About 95% of the observation hot ct neaa © 28D (1 © 2a}, About 99°% of the obvervations will he horween mean & 38D Gey £ ty WWW .[Link] Business Stati Standard normal probability distribution If X is a normally distributed random yariable u and @ are respectively ts mean and standard X=u standard normal random variable. Special table called table of reo areas under normal curve is available to determine probabilities thatthe random variables lies in a given range of values of the variables. Using the table, we can determine the probability forX, taking a valueless than x (X < x} and also for a given probability we determine the value x such thatX 182) = PZ > 3) = $-P(0Z) = Bor PO80) = 05S Let Z, and Z, be the values of Z. where X = 3 and X = 80 respectively Then P(Z,< Z<0)= 40 From tables, Z, “= —1.28 neatly wos hey 1.28 or Mp = 120 Also P< ZEZS = AS Z, = 104 orm Distiton @ Seo wn = tote w= 188 way = 164 2-30 ee o = 399 1712 Fron (i, = 4128x1712 = Stocaty When X= 45,2 = 28 When X = 60, Z JP(a5 2) = 5-P(Z>2) $= 477 = 0.023 2 0 When X= 150,2 = WWW .[Link] ie Business Statisties P(Z<~1) = P@>1) = S-P(Z>1) = $= 0341 = 0.199 Example 13: The customer accounts of a certain departmental store have ‘an average balance of Rs.120 and a standard deviation of Rs.40. Assuming that the account balances are normally distributed, i) what proportion of accounts is over Rs.150)? Gi) what proportion of accounts is hetween Rs.100 and Rs.1507 Gi) what proportion of accounts is between Rs60 and RSA? Sotution : @ = 120, 0 = 40 z 10-120 _ When jal =4 X= is, 2 = He P(X > 130) = FIZ > «19) > 1= PZ <5) - 5-027 0.2260 fe, 226017 cl accounts wil have balance ver Rs 150. (i) 2 Wand = 100 2 nA) S100, 120150 a 25 ens e 480 = 120 When X= 1, 2-85. PAM EX 15M eP(-S <7 < 75) S562 <0) = Pie Z~ 15) + BT = an) 075 PO 48) = P(Z>15)_ so = a r=16 20 = 5-P@%<16 = 5-045 = 0055 « Proportion of jabs which takes more than 48 minutes = 0.55, GHP 5 < X < 48)=P(-1< 2 < 16) = P(-1<7< 0) +PO

You might also like