You are on page 1of 72
CHAPTER Standard Distributions ZU4o Syllabus Topic : Random Variable 1.1 Random Variable What is Random Variable? Define Random variable with example. 3 fandom Variable (R.V.) * Experiment is any process that generates outcomes. In an experiment we are interested in a Particular outcome but instead of that if we concentrate on the process of assigning numerical values to experimental outcomes. Such a process is called a random variable. For example, in a tossing of two coins simultaneously, we may have interest in number of heads oblained in an outcome instead of sequence of heads and tails. There is a rule to assign a number ‘0 an outcome that is called a random variable. * — Arandom variable is a numerical description of the outcome of an experiment. * Definition: A random variable is a function from a sample space into the real numbers. In this example X = number of heads, X is called a random variable. The values that X takes it ‘depend on outcomes of the experiment. Q = (HH,HT,TH,TT) | nH | ar | tH [tr x|2 7 1 lo ‘The range of random variable X is (0, 1,2) LE stat. methods & Test of Hypo (B.Se-Comp-MU) _1-2 Standard Distributions I Examples of Random Variable = Experiment Random Variable Range of random No. variable i Family with 3 children Number of girls in a family 0,1,2,3 2 Inspect a shipment of 50 | Number of defective items O12. coo. 50 items 3 Picture tube in a new | Lifetime of the picture tube 0 0 forall x in sample space Gi) Z, P(w)= 1 Then the function P is called the probability mass function (p.m.f.) of random variable X. 1.2.2 Probability Distribution © Definition : Let X be discrete random variable with probability P (x). The ordered pair (x,, p(X;)) is usually depicted in tabular form as follows: x X X. | Xs Xq_| Total PLEX =x] | P(x,) | P(x) | POs) P(x,) | . ‘The tabular form of X and p(x) is called probability distribution of X uppose a family of 3 children. Let X is number of gifs in a family. Find probability distribution of X. Consider a family with 3 children. We define sample space as © = (BBB, BBG, BGB, GBB, BGG, GBG, GGB, GGG) Let X is number of girls in a family. X = 01,23 The p.m. of X is 1 P(X=0) = P(BBB)=g 11.1.3 P(X=1) = P(BBGorBGB or GBB)=g+g+§ =g 1.1.1.3 P(X=2) = P(BGG orGBG or GGB)=3 +g +3=3 1 P(X=3) = P(GGG)=_ Stat. Methods & Test. of H The probability distribution of X is x o{1]2| 3 | Total ey{1}3/3])1] 1 POD le la la la Note : Functions of Random Variable if X is @ random variable and Y=g(X), then Y itself is a random variable and probability mass function of Y is same as p.m. of X Ex. 1.2.2: For the following probability distribution of X x ij2 {3 |als PX=x) | k | 3k | 2k | k | 5k () Find value of k (i) Find P(X> 3), P(x <4) (iii) Find probability distribution of Y = 2X — 1 i) As P(x) is ap.mf., it should satisfy the properties of p.m.f. which are (a) P(%;) > 0 forall x in sample space & E,PmK)=1 * k>Oand 2k=1>k=4 12 P(X23) = P(X=3)+P(X=4)+P(X=5)= 8k= ab=3 iii) PX <4) = PO = 1) + PO =2)4 PK =3)=k + 3k +2k = 6k = 6b = iv) Y=2X-1, which is function of X Y takes values — 1, 3, 5,7,9 As Y is function of X, the p.m.f. of Y is same as p.m.f. of X - Probability distribution of Y is i 2 ¥Y- 4-1/3 5 7 9 aol 1 | 3 2 1/3 PY») | [72 [72 | 79 [72 Syllabus Topic : C.D.F for Discrete Random Variable 1.2.3. Cumulative Distribution Function (C.D.F.) or Distribution Function ( D.F.) |. _Define Cumulative Distribution Function or Distribution Function. | Definition : Let X be discrete random variable with P.m.f. P(x). The cumulative distributi function of X denoted by F(X) is defined at Probability accur “ -_ mulate upto point X i.e. FT stat. Methods & Test. of Hypo.(B.Sc.- Comp.-MU) _ 1-5 ‘Standard Distributions ne FX) = P(Xa)=1-P (X4) 5.P(X23) 6. PIX=5/K23) 7. P(X<5/X 22) Soln. : Since FOX)= Zh _ g Pw) FH.) = ait Pw Pra = FOR) - FX) 1. The p.m. of X is given by x li [2 3 aoe see |e Fx) | 0.2 | 0.4 0.45 0.6 | 0.68 | 0.82 | 0.9 | 1.00 P(X) | 0.2 | 0.4-0.2=0.2 | 0.45 -0.4 = 0.05 | 0.15 | 0.08 | 0.14 | 0.08 | 0.1 Pa) = FU)y=02 PQ) = F2)-F(l)=04~-0.2=0.2 PG) = F(3)—F(2)=0.45 - 0.4 = 0.06........ 2 P(24) = 1-F(4)=1-06=04 By property 5 OrP(X>4) = P(S)+P(6) + PC) + P(8) = 0.08 + 0.14 +.0.08 +0.1=0.4 . P(X23) = 1-FG)+P@)=1-045+0.05 =0.60 [EF stat. Methods & Test. of Hypo.(B.Se.- Com 17 Standard Distributions By property 6 Or P(X>3)= P(3) + P(4) + P(5) + P(6) + P(7) + P(8) = 0.05 +0.15 +0.08 + 0.14 +.0.08 + 0.1 = 0.60 6. Tofind P(X =5/X >3) Recall P(A/B) = eae Letus define A = (X=5)andB=(X>3) -p(A) _ PAOB) 2G) = SH AndAQB = (X=5)0(X23)=(K=5)O(X=3,4,5, 6,7, 8) =(K=5) P(X=5/K23) = Pea = 998 <0.1333 _ PX <5X22)_P2SX<5) _F(S)—FQ)-PS)+PQ)_04 7. POR 4s Var) = Eig (Kw POX) = EX’) -(E00)'=5.0-(2.0)'=5-4=10 1i) Mean and variance of profit h(x) BY using properties, E (h(X)) = E (800X - 900) = 800E(X) ~ 900 = 800 x 2 - 900 = 700 Var (h(X)) = Var (800X ~ 900) = 800" Var(X) = 640000 x 1 = 640000 SD(h(X) = «/Var (h(X) = 640000 = 800 oR x Gaye 2 3 | Tom | ¥=h00 | ~900 | — 100 | 700} 1500 | Px | ot!) 02] 03] 04 1| YP) | -90) -29 210 600 700 | ¥"Pex) | s1000 | 2000 | 147000 | 900000 | 1130000 | As is function of X so pms. of Y is same as pam. of X “ EY) = Ly pxy=700 EY) ayy? 130000 - (700)? 1130000 ~ 490000 = 640000 War Y) = V540005 = 800 Solved Examples based on Discrete Random Variable Ex.1.2.9: Let X represents the difference between the nd the number of tails Ghlsined when a coin is tossed 3 tes. Find Probability dite 2 Distribution of X ii) Median of X ‘rbution of X i) Cumulative number of heads ay Soin. : In an experiment of tossing of coin 3 times sample space ig Q = (TTTTTHTHTTT. THE.HTHHETHHH) Let X isthe difference between the number of heals and the number i wits X takes values ~3,—1, 1,2,3 pm. of X is P(X=-3) = PIX=-1) = POTH.THT.HTT) = 3=0.375 P(X=1) = PCTHH.HHT,HTH) =2 = 0375 P(X=3) = (sta. Methods & Test. of Hypo.(B.Sc.- Comp.-MU) 1-13 Standard Distributions 00 SE Probability Distribution of X x -3 | -1 | 1 | 3 | Tota PR =x) | 0.125 | 0.375 | 0.375 | 0.125 | 1 Fx) | o.125 Jos [oss] 1 Median divides distribution in two halves. Let M denote median, according to definition. P(KS M) = 05 F(M) = 05 .M=-1 Ex. 1.2.10: Verify whether the following can be p.m. for the given values of X. ) x 2 |3 [4 |s |e PK=x) | 0.15 | 0.25 | 02 | 0.1 | 0.3 il) x -1{o [4 P(K=x) | 0.2 | -0.1 | 03 | 03 | 03 (ii) PX)= J forX=0,1,2,3,4 3 (iv) PR) forx= 1,2, 3,4, 5,6 Soln. : P.m.f. should satisfy following condition (a) P(X)20 Vx (bo) ZP(X)=1 (i) P(X) 20 forx =2,3, 4,5, 6 ZPR) = 0.15 +0.25+0.2+0.1 +0. It satisfies both conditions so it is p.m.f. =-0.1<0 +. P(X) is not p.m. wd dtetel abet 2 poo loapms <0. P(X)is not pm.f Ex. 1.2.11 : Determine k such that the following are p.m..s () PEQ= KACEY -X=0,1,2,3,4,5 ox W) PRH=Yp X= 01.2.3 Standard Distrburt, Soin. @ DPR) = ZP(K) = ke $ AS ('Cy#"C, °C, +...4°C,)=2" (i) LPH = 1 _ EP(K) aks ake a4 BE 1 3 ke 5 Ex.1.2.12: Following the cumulative distribution function of a discrete random variable X. x {rtelala/s]e{7 F(X) | 0.09 | 0.23 | 0.95 | 0.49 | 0.71 | 0.89 | 1.00 Find i) pamJ. of X i) Mean ii) Standard Deviation iv) P( 2 2) Soln. : Forty. X, pamf. of X PK) = F(X) - FX) xii a a 4 5 6 7 Total FO) [ooo] 023 og art 089 1.00 P0009 023-000 0.14 049-035 =0:6 071-048 =022/089~071 =018| 100-089=0.11|1 P(X) | 0.09 028 0.36 056 1.10 1.08 om 424 eppx)|009/ 086 108 224 550 6a 539 [ai Mean = o)= Z| X P(X) =4.24 Var(X) = E(X)-(E(X))' = 21.34 - (4.24? = 3.3604 ‘Standard Deviation = $.D,(X) = [Var (X) = 33624 = 1.8337 P(2SXS6) = P(K=2)4+P(K=3)4P(X=4) + P(X =5)+P(K =6) = 0.1440.12+0.144022+0.18=08 ForPQX=4/X22)A = (X=2) B=(X22)=(X=2,3,4,5,6,7) ANB= (X=2) (X=4K22) = Pap) =P St: . 0.14 © 0.14 +0.12+0.14+0.22+0.18+0.1 Ex. 12.13: A box contain § red and 5 green balls. Two balls are drawn at random, they are ofthe ‘same colour, then you win Rs 105 ; if they are of different colours, then they lose Rs 100. Find the expected value and variance of the amount you win. BB sat. Methods & Test_of Hypo Comp.-MU) _1-15 Standard Distributions Soin. : Let X be amount to win P(X =200) = P (both balls are of same colour) = P (Both balls are either red or green) = P (both balls are red) + P(both balls are green) ea cS 10 10 _20 = 7 Teas tas 45745 P(X =-100) = P (both balls are of different colour) 5 Cc = P one ball sof red color and one ball is of geen color) = w+ = 5 2 _ = 108 22 20 _ 2100 | / 2500) _ 2100-2500 EX) = EX P(X) = 105 x35 + (- 100) x 3 7 ¢- 25)" a5 =p =- 8.889 Rs, 2) = Ex? payers? 22 2 25 _ 220500, 2sco00) EX’) = Ix POX) = (105)"x gs + (- 100)'x 35 = Ge +( 3 470500 45 = 10455.56 Var (X) E(X) — (B(X))" = 10455,56 - (~ 8.889)"= 10376.54 Syllabus Topic : Continuous Random Variable 1.3 Continuous Random Variable |. Explain Continuous Random Variable. There are three types of sample spaces: finite, countably infinite and uncountably infinite. A sample space which is finite or countably infinite called countable. If the sample space is not countable or uncountably then it is called continuous. A continuous sample space is subset of real line R. Ilustrations of Continuous sample space (Suppose in an experiment, the average inter-arrival time between two train is 10 mins. The sample space for arrival of next train will be an interval in the neighbourhood of 10 min such as. Q={8, 18) min (ii) Suppose in an experiment, life of an electric tube in hours is recorded. The sample space is = {0,7000) hours. Recall that range set of continuous random variable is uncountably infinite. Illustration of continuous random variable (4) Weight of an oil bag having capacity of 1 Kg (b) Time to copy file from pen drive to hard disk (©) Life in hours of an electric component 7 BP sta Methods & Test. of Hypo. 1-16 Standard Distribution, ‘Syllabus Topic : Probability Density Function (p.d.f) 1.3.1. Probability Density Function (p.d.t.) lQ. Define p.df. _ ats oB For discrete random variable, using p.m.f. we get probability of random variable, while for continuous random variable probability mass is not attached to any particular value. It is attached to an interval. The probability attached to interval is called density function. Definition of Probability Density Function (p.d.f.) Let X be continuous random variable. Function f(x) defined for all X € (~ ©, %) is called probability density function (p.d.f.) if @) f(x) 20 - 0 oo “15 64 Hence f(x) = 75 ii) f(x) = S52 implies 2 0 BP stat Methods & Test. of Hypo.(B.Sc.- Comp.-MU)_1-18 Standard Distribute, J tyax = nome c= en Therefore f(x) = +—> m1+x Ex.1.3.3: Let X be continuous random variable with p.d.f fx) = fx+e forosxs2 =0 otherwise Find i)e ii) P(11.5/1< X <2) Soln. : As f(x) is p.d.f. of X it satisfies both conditions of p.d.f. 2 J to ax ey i) Jax) dx= 1 1 2 1 Ix IGxre a = 13 [gF+ex| =1 0 [2+ 2] @ 1 ee ee For]=H[( Doan] 2 Gi) PUcxX<2) = Joo ax = f+ 1) ax 7 i i 1 +G+ ] = 3 = 0.375 (iii) P(X > 1.5/1 1.5) oe (1 1.5/1a) = P(X 2a) =1-P (X0=1-F) where F is the distribution function of the component lifetime. More specific, reliability is the probability that a product or part will operate properly for a specified period of time under the operating conditions (such as temperature, volt, e7c.) without failure, In other words, reliability may be used as a measure of the system’s success in providing its function Properly. Reliability is one of the quality characteristics that consumers require from the manufacturer of products. Ex. 1.3.4: Let X be a continuous random variable whose probability density function is 2 10 =4 02 x 2, ox, Fa) = PXsH= J yay = Joay+ Jays May <04140- ° -« o + Cumulative Distribution Function of X is FX) x2 Standard Distributions Ex. 1.3.5: If X's continuous random variable with distribution function FX)= 0 X<-4 =n ~11 Ts f@) = 40 -16x<1 =0 otherwise ii) P(24] Syllabus Topic : Variance of Continuous Random Variable Aaa Mean and Variance of Continuous Random Variable (Q. Explain mean and variance of continuous random variable, ‘As we have seen how to find mean and variance of discrete random variable , similar way itis defined for continuous random variable. The summation sign is used in case of discrete r.v.is replaced by integration in case of continuous r.v. Definition Let X be continuous random variable with pad. f(x), then mean of X or expectation of X denoted by E(X) orp is defined as BX) = p= J xfixydx And Variance of X denoted by Var (X) or is defined as. Var(X) = EUX-wil= J (X-p) A) dx = EX')- GO) where EX?) =f x? f(x) dx Ex. 1.3.7: If Xis continuous random Variable with pdf. f(x) =f 1 10) ii) = [e°-e'}= 10 one (ii) P(S1 FX) = 1 Distribution function of X is FX) = 0 x<0 = x' (3-2) 01 Gi) PEK <4) = FU= Syllabus Topic : Binomial Distribution 1.4 Binor BP stat. Methods & Test. of Hypo.(B.Sc.- Comp.-MU) _ 1-27 Standard Distributions fx) = PK =n=()pta =p) ‘An example of the graph of the pmf is given in Fig. 1.4.1. *, x=0,1,. n= 10, p=0.7 0.25: 0.2 0.15: 4 0.05: o 7 2 3 4 5 6 7 8 8 10 Fig. 1.4.1 : The pmf of the Bin(10, 0.7)-distribution Alternatively Binomial distribution can_be defined as A random variable X is said to follow binomial distribution with parameters n and p if P (X)= "C, p'q’ * where x =0, 1, 2,3....n, p is the probability of success and q is the probability of failure and q = 1 — p and p(x) is called the probability mass function. Note A For a binomial distribution 2 p(x) =1 x=0 In a binomial distribution 1) n, the number of trials is finite. 2) each trial has two possible outcomes called success and failure. 3) all the trials are independent 4) p and hence q is constant for all the trials. —— Syllabus Topic : Properties of Binomial Distribution 1.4.1. Properties of Binomial Distribution @. —_ Explain Properties of Binomial Distribution. 1. IfX ~B(n, p), Mean of X is mp 2. Variance of X is npq 3. Additive Property : If X ~ B(n,, p) and Y ~ B(n,, p) are independent X+Y~B(n, +n, p) random variable then stat Methods & Test. ot Hypo (8.Se- Comp: MU) _1-28 Standard Distributions Sat Methods & Test of Hypo 0 5 Ot Additional Information on Properties of Binomial Distribution ‘The expectation is E(X) = np. This is a quite intuitive result. The expected number of successes (heads) in n coin tosses is np, if denotes the probability of success in any one toss. To prove this, one could simply evaluate the sum a x x(*)pta Pas x=0 But X can be viewed as a sum X = X, + .... + X, of n independent Ber(p) random variables, where X; indicates whether the i-th toss is a success or not, i =I,....n. Also we will prove that the ‘expectation of such a sum is the sum of the expectation, therefore, EX = E(X,+....+X)= X\+.... HEX, = p+. we =np ‘n times Numerical Problems Ex. 1.4.1: Itis observed that, if student works hard then chances of Oy Q) passing a exam is very high Le. 80%. A random sample of such 10 student were selected. What is the chance that (i) No ‘student will pass exam (ii i) 3 students will pass exam (ii) More than or equal to 8 students will Pass exam (iv) All of them will pass exam, Here passing a student is Bernoulli trial, X + Result of a student, P> Probability of passing a exam, P = 80% = 0.80 “+ q — Probability of failing a exam = 1-p=1-0.80=0.20 Pmf of binomial distribution PXX=x] = 'C,Ptq?* Heren — No. of students n= 10 Probability that no student will pass exam ie. x = 0 PIX=0] = "Cop" g'* = "C, (0,80)° (0,20)""-° = 11-020)" 2%, =1,(080)°= 1) = 0.20)" Probability that 3 students will pss exam, Here 9 2 PIX=3] = °C, (0,809 (0.20)!"-3_ 10! 10-9 10-98-71 2s = 1Gx2x1 08) (020)'=10-3-43 080.080 x08 (0207 120 x 0.64 x 0.8 x (0,20)" =10,x=3 (0.80) (0.20)? = 61.44 x (0,20) 0.00078 BF sat Methods & Test. of Hypo.(B.Sc.- Comp.-MU) _ 1-29 Standard Distributions I (3) Probability that more than 8 students will pass exam. PIX28) = PIX=8]+P[X=9]+P[X=10] "°C, (0.80)* (0.20) + '°C, (0.80)° (0.20)! + "°C, (0.80)"° (0.20)° (4) Allof them will pass exam. P[X=10] = "C,P*q"~*="°C,, (0.80)'° (0.20)’= (0.80)'° Ex. 1.42: Vijay has started new retail out let in mid of the market. In market there is business and competition. Therefore survival of a new outlet is very rare chance of survival is almost 5%. Vijay has started such 7 new retail outlet. Find out the probability that (1) At least 3 shops will survive. (2) Exactly 5 shops will survive. soln. : Here success or failure of new step is Bernoulii variable s. X > survival of new shop i No. of shops started = 7 Z2t4+o Probability of survival of outlet = 5% = 0.05 Probability of non survival of shop = 1 - 0.05 = (1) Probability that at least 3 shops will survive PIX23] = 1-P[X <3] =1-(P[x=2]+P[x=1]+Pix=0]] 1~['C, (0.05)? (0.95)° + 7C, (0.05)' (0.95)° + "Cy (0.05)° (0.95)'} sor x 0.0025 x (0.95)° +7 x 0.05 x (0.95)° + 1-1 95)’ | 7 " 1.95, . " 0 1 ~ [21 x 0.0025 x (0.95)° + 0.35 x (0.95)° + (0.95)'] 1 — [0.0406 + 0.257+0.6983] = I — 0.9962 = 0.00376 (2) Exactly 5 shops will survive "Cc, p*q"* ="C, (0.05)° (0.95) " 7x6 2x1 P[X =5] sa 0.05)° (0.957 x 0.05)° (0.95)* 7 x 3 x (0.05)° (0.95)" = 5.922E — 06 = 0.000059 = 0 Ex. 1.4.3: Chances that a student of Mumbai University clears UPSC exam is 70%. If 20 Students appear for the exam. What is the chance that more than 3 students will clear the exam. Soin. : Let X denote result of UPSC exam. P_ — Probability of Pass (clearing UPSC exam) — 10% = 0.70 4 — Probability of failure of exam = No. of students = 20 Probability that more than 3 students will pass exam PIX>3]) = 1-P[X<3]=1-[P(K=3)+P(X=2)+P(K = 1) +P(K=0)] -P=1-0.70=0.30 [FF stat. Methods & Test. of Hypo.(B.Sc.- Comp.-MU) _1-30 Standard Distrib, 1-[°C, (0.70)* (0.30)"” + °C, (0.70)" (0.30)'* + °C, (0.70)' (0.30)' +™Cq (0.70)° (0.30)"} Ex. 1.4.4: Ina cricket match between India and Pakistan, India’s batsman Mr. Dhoni is on strike, Lag) over is going on. From the past record of this batsman it is estimated that chance of lting «| six is very high that is 90%. What is the chance hit Mr. Dhoni will hit at least 4 sixes. Soin. : Heren = 6 (as6 balls in a over) P = 0.90 Probability of Lifting six | q = 1-P=1-090=0.10 Probability of not hiting six +. Probability that Mr. Dhoni will hit at least 4 sixes. PIX24] = P[X=4]+P[X=5]+P[X=6] = °C, 0.90)" (0.10)" + §C, (0.90)§ (0.10)' + °C, (0.90)* 0.10)" 6x5 = cq (0.90)" (0.10)" + 6 x (0.90)* (0.10) + 1 (0.90)°1 = 0.0984 + 0.3542 + 0.5314 = 0.984 = 98% Problems based on Binomial Distribution Ex. 1.4.5: The mean and variance of a binomial variate are 12 and 6. Find )P(X=0), ii) P(X21) Soin. : Mean = w= np=12 Variance = 0° =npq=6 mq _ 81 1 np = 1672 ke. q=a P= mp = 12 ien=24 i) P(X=0) = % i Q-@" ii) P(X21) = 1-P(X <1) =1-PK=0) Ex. 1.4.6 : Six dice are thrown 720 tir Pee Imes. How many times do you expect at least 3 dice to show a Soin. : Here on = 6,N=720 P(x23) = °C, pig’ Let P be the probability of getting 4 or 6 with 1 dice 6 OO cerca om 3 “3 (3) +c, GY 2) 46 number of times = 720 0.3196 = 230.1934 eo ” A basket contains 20 toys Basket. Find the brobety eee r8N968 3 oranges are dawn a random 900d oranges. Stout of 8 i) exacty 2i) at east 2) at nomenon Soin. : Let P be th a , . ‘ e ry Of getting a good orange ie,P = che T= 0.84 (PRK 2)="C, (0.8)°(0.2)' = 0384 Gi) P(X22)=P(2)+ Pay = "C8 (0.2)' + °c, (0.8)' 0.2)" 1.384 + 0.512 = 0,896 (ii) PXS2) = PCO) +P(1)+ P(2) = °C, (0.8) (0:2)! °C, 0.8)! (0.2)? + °c, 0.8702)! = 0.2 + 0.096 + 0.384 = 0.68 Ex. 1.4.8: In a sampling a large number of parts Manufactured by a machine, the mean number of defective in a sample of 20 is 2. Out of 1000 such ‘samples how many would expected to Ex. 1.4. -08=0.2 Contain atleast 3 defective parts. Soin. : n = 20np=2 ; 1 2 jeep = 79 q= 1-p=79 P(X23) = 1-p(x<3) =1-p(x=0, 1,2)=0,323 Number of samples having at least 3 defective parts = 0.323 x 1000 = 323 Syllabus Topic : Normal Distribution 1.5 Normal or Gaussian Distribution lO. Whati ion? Or Define N ai ‘A random variable has a normal distribution with parameters j2 and o° if its density function f is rat given by s [EP stat. mothods & Tet of Hypo.(B.Sc.- Comp.-MU) _1-32 Standard Distribution, ——Sy We write X ~ N(u, o°), It means X follows normal distribution with mean y. and variance g' Parameters 1 and 0” turn out to be the expectation and variance of the distribution, respectively. Ipsg and o = | then fx) = Neos) and the distribution is known as a standard normal distribution. The cdf of this latter distribution is often denoted by, and is tabulated. Neo.) The probability densities for three different normal distributions have been depicted. 3 2 4 0 1 2 3 « = Fig. 1.5.1 Net) Syllabus Topic : Properties of Normal Distribution 1.5.1 Properties of the Normal Distribution |. _ Explain Properties of Normal distribution. 1 IX ~N (u, 03), then X=! N , 1) ‘Thus by subtracting the mean and dividing by the standard deviation we obtain a standard normal distribution. This procedure is called standardisation. ‘Thus Z has a standard normal distribution. Standardisation enables us to express the cdf of any ‘normal distribution in terms of the cdf of the standard normal distribution. This is the reason why only the table for the standard normal distribution is included in the appendix. 2. A trivial rewriting of the standardization formula gives the following important result : If X~N(q, 0°), then X = +02, with Z~N(O, 1). In other words, any Gaussian (normal) random variable can be viewed as a so-called affine (linear +constant) transformation of a standard normal random variable. 3. EX = LL. i.e. mean of normal distribution i M 4. Var(X) = 0”. This is a bit more involved. First, write X = + 0Z, with Z standard normal. Then, Var(X) = Var (+ 0Z) = 0" Var (Z) EZ = Standard Distributions Stat. Methods & Test. of Hypo.(B.Sc.- Com since the last integrand is the pdf of the standard normal distribution. ‘A bell shaped symmetric curve ‘Symmetric, arithmetic mean, median and mode coincide. 1.6 Standard Normal Distribution a. _ Define Normal distribution with its properties. @. _ Explain Properties of Normal distribution. Normal Distribution Graphs of Normal Probability Distribution sr. Properties of a normal curve Figure No. 1 The curve is bell-shaped with highest point over the mean }. 2. ‘The curve is symmetrical about a vertical line through 3. The curve approaches the horizontal axis but never touches or crosses it. 4 ‘The inflection (transition) points between cupping up and down occur at # - 6 and pte. 5. The total area under the curve is 1. 6 ‘The mean p and standard deviation 6 control the shape of the normal curve. e. The curves on the right show examples of normal curves of different height. 6 A I. : 5 c 3 40 60 «0070 Fig. 1.6.2 FT ow. ( mp.-MU) Standard Distribution, Empirical Ri For a distribution that is symmetrical and bell-shaped 1 Approximately 68% of the data values will lie within one standard de: Viation on each side of thy ‘mean, Approximately 95% of the data values will lie within two standard mean. deviation on each side ofthe Fig. 1.6.3 | 1.6.1 Standard Normal Variate Q._ What is Standard Normal Variate or Define Standardization. Values on any normal distribution x can be transformed using the formula Z = mm This is usually referred to as a z-score or z value Percents (areas under the curve) can be derived using the standard normal. Before working with the conversions for the x normal distribution, you must be able fo work with the standard normal distribution table (Table in the appendix or the pull-out card). The table shows negative values which are to the left of the mean of 0. The other side of the table shows positive values which are to | the right of the mean of 0. Fig. 1.6.4 ‘Standard Distributions 1.6.2 Additive Property of Normal Distribution la, __ State Additive property of Normal Distribution Statement Suppose X~ NO, a) and Y ~ N(j,, 6, ).IfX and Y independent random variables then X+Y-N(Hi+Hy 9, +9,) Generalisation TEX, Xsyoe-Xp are independent random variables such that X, ~ N(H, 0, ) then Ex,-N(Zim,20,) Corollary 1X1, X, (@ 2 X,~N (ap, no") and ) X-N (2) and X, are independent identical distribution (iid) N(q,, 0”) variables then 1.6.3 Normal Approximation to Binomial Distribution l@. State the De Moivre's theorem on Normal approximation to Binomial Distribution De Moivre’s Theorem : If X follows binomial distribution with parameters n and p then probability distribution of X tends to normal with mean np and variance npq as n- © (n is large) |. X=mp ie. Faprteaemees 1.6.4 Finding Areas under the Standard Normal Distribution Curve 1.6.4.1. Area under Standard Normal Curve For the solution of problems using the standard normal distribution, a four-step procedure is recommended with the use of the Procedure Table shown. Step1: Draw a picture. Step2: Shade the area desired. Step3: Find the correct figure in the following Procedure Table . Step4: Follow the directions given in the appropriate block of the Procedure Table to get the desired area. TABLE A : Areas of a Standard Normal Distribution The table entries represent the area under the standard normal curve from 0 to the specified value ofa 0.0 0120 0239 0279 0319 0359 0.1 .0398 0478 .0517 0557 0636 .0675 Ona 0753 0.2 0793 0871 0910 0948 1026 1064 1103 “1141 03 21179 1255 1293 -1331 -1406 1443 -1480 1517 0.4 1554 1628 1664 1700 772 1808 1844 -1879 0.5 1915 1985 2019 2054 2123 2157 2190 2224 0.6 2257 2324 .2357 2389 2454 2486 2517 2549 0.7 2580 2642 2673 2104 2764 2794 2823 2852 08 2881 -2939 2967 -2995 3051 3078 3106 3133 09 3159 3212 -3238 3264 3315 -3340 3365 3389 1.0 3413 3461 3485 3508 3554 3577 3599 3621 -3643 -3686 -3708 3729 3770 -3790 -3810 .3830 12 3849 -3888 -3907 -3925 -3962 -3980 -3997 4015 13 4032 4131 4147 4162 AITT 14 4192 4222 4236 4251 4279 4292 15 4332 4357 4370 4382 4306 4319 4418 4429 1.6 4452 4474 4495 4441 ASIS -4525 4535 17 4554 4873 4582 4591 4608 4545 4616 4625 4633 18 4641 4656 4671 4686 19 4713 4726 4732 4738 4706 4750 2.0 A712 4783 A788 4793 4756 4761 4767 4803 2.1 | 4821 4830 4834 4838 4846 -4808 4812 4817 4850 4854 4857 2.2 | 4861 | 4864 | .4868 | 4871 | .4875 | .4878 | .4881 | 4884 | 4887 A890 2.3 | 4893 | 4896 | .4898 | .4901 | .4904 | .4906 | .4909 | .4911 | .4913 | 4916 2.4 | 4918 | .4920 | .4922 | .4925 | .4927 | .4929 | 4931 | .4932 | .4934 | .4936 25 | 4938 | .4940 | 4941 | 4943 | 4945 | .4946 | .4948 | .4949 | .4951 | 4952 2.6 | 4953 | .4955 | .4956 | .4957 | .4959 | .4960 | .4961 | .4962 | 4963 | 4964 2.7 | 4965 | .4966 | .4967 | .4968 | .4969 | .4970 | .4971 | .4972 | .4973 | .4974 2.8 | 4974 | .4975 | .4976 | .4977 | .4977 | .4978 | .4979 | .4979 | .4980 | 4981 2.9 | 4981 | .4982 | .4982 | 4983 | .4984 | 4984 | .4985 | .4985 | .4986 | 4986 3.0 | 4987 | .4987 | .4987 | 4988 | 498s | 4989 | .4989 | .4989 | .4990 | .4990 3.1 | 4990 | .4991 | .4991 | .4991 | 4992 | .4992 | .4992 | .4992 | .4993 | .4993 3.2 | .4993 | .4993 | .4994 | .4994 | .4994 | .4994 | .4994 | .4995 | 4995 | 4995 3.3 | .4995 | .4995 | .4995 | 4996 | .4996 | .4996 | .4996 | .4996 | .4996 | .4997 3.4 | 4997 | .4997 | .4997 | .4997 | .4997 | .4997 | .4997 | .4997 | .4997 | 4998 3.5 | 4998 | .4998 | .4998 | .4998 | .4998 | .4998 | .4998 | .4998 | .4998 | .4998 3.6 | .4998 | .4998 | .4998 | .4999 | .4999 | 4999 | .4999 | .4999 | .4999 | .4999 For values of z greater than or equal to 3.70, use 0.4999 to approximate the shaded area under the standard normal curve. Case1: Area or Probability Between zero to any number (a) P 1.53) = P(Z>0)-PO2.56) due to symmetry P(Z > 0) PO -0.45) = P(Z<0.45) due to symmetry = P(Z<0)+P(0b) = 0.28 findd 0.5-POc) = 6g find ¢ 0s +PO 125) For 125, Z score is 2 = Xa _ 25-100 _| -_— “ P(X>125) = PZ>1)=PZ>0)-PO 3300) eee os EP stat. methods & Test. of Hypo.(B.Se.- Comp MU) 1-42 Standard Distribution, 3300-4400 __ 1.77 (round to 2 decimals since the normal table only gives 2 decimals. So, P(X > 3300) = P(z>-1.77)=P(Z< 1.77) due to symmetry 5+P(01.2)=0.5-PO2) = P(r” 25) =P(z>08) = 05-PO) Find the probability thatthe weight is between 20 kg and 27 kg. (c) Find the probability that the weight is more than 29 kg, (8) If 50 goats are randomly selected, about how many would you expect to weight less than 22 kg, Find the percent first and then find frequency BP sa Methods & Test. of Hypo. B. Comp. MU) 1-44 Standard Di tributions Sotn. @ Ax<2H = P(r< 5) = pez <-0.67) = P(Z >0.67) due to symmetry = 0.5-P029) = = 05-PO1)=0.5 0.3413 = 0.1587 No. Of goats weight less than 22kg = 50x P(X <22)=50 x 0.1587) = 7.933 = 8 goats 4 Fig. P.1.6.3(c) FEF srat. Methods & Test of Hypo.(8 Comp-MU) _1-45 Standard Distributions 5-41.84: The tim takan fo assonbio cari a random vale having aroma dvi Ot hours standard deviation of 2 hours. What is the probabilly that a car can be assembled at this plant in a period of time “ 1a) less than 19.5 hours? —_b) between 22 and 25 hours? Soin. : Let X : Time taken to assemble a car. X ~ N(20, 2) a) PX< 19.5) = P(r<2S= ) = P(Z<-0.25)=P(Z>0.25) due to symmetry = 0.5-P(080) Zs 0-70 P(x>go) = PZ>1) =05-PO 60) Sc.- Comp.-MU)__1-46 tat. Methods & Test. of Hypo. 60-70 Zz 10 P(X>60) = P(Z>-1)=P(Za) = 0.10 a, = 2 = Ppz>a) = 0.100 05-PO0.45) = 0.3264 Directly from table of Right tail 045 Fig. 1.6.24 Ab) P(Z< 1.24) = 1-P(z> 1.24) = 1 - 0.1075 = 0.8925 (©) P(Z<-0.25) = P(Z>0.25) due to symmetry = 0.4023 0.25 0.25 Fig. 1.6.26 -@) PZ>- 1.34) = P(Z< 1.34) due to symmetry —PZ > 1.34) = 1 ~0.0901= 0.9199 Fig. 1.6.27 0.42 2.07 Fig. 1.6.28 () P(-0.25 1.07)] = 1 - [ P(Z> 0.25) + P(Z > 1.07)] = 1~[0.4404 + 0.1660] = 0.3936 0.25 Fig. 1.6.29 Ex,1.6.6: The annual salaries of employees in a large company are approximately normally distributed (a) b) with a mean of Rs.50,000 and a standard deviation of Rs.20,000. a) What percent of people earn less than Rs. 40,000? b) What percent of people earn between Rs. 45,000 and Rs. 65,000? c) What percent of people earn more than Rs. 70,0007 Here X is the annual salaries of employees in a large company. X ~ N(50000, 20000) P(X < 40000) 40000 — 50000 _ —— 20000 =-05 P(Z<-0.5) P(Z>0.5) due to symmetry 0.3085 (Using Right tail table) 30.85% earn less than Rs. 40,000. P (45000 < X < 65000) For X = 45000,z =—0.25 and for x = 65000, z = 0.75 05 0s Fig. P.1.6.6 >. Standard Distri Pout Methods & Test. of Hypo.(B.Sc.-Comp.-MU) _ 1-50 tht, P(45000 < X < 65000) = P(-0.250.75)] = 1-[P(Z>0.25) +P(Z>0.75)] due to symmetry = 1~[0.4013 + 0.2266] = 1-0.6279=0.3721 0.25 0.75 Fig. P.1.6.6(a) 37.21% earn between Rs. 45,000 and Rs. 65,000. ce) P(X > 70000) Forx = 70000, z = 1 P(X > 70000) = P(Z> 1) = 0.1587 15.87% earn more than Rs, 70,000. 1 Fig. P.1.6.6(b) 'ean 1 = 30 and standard deviation o = 4. Find a) P(x<40) b) P(x > 21) ©) P(30 25)= 1 ~ 0.0062 = 0.9938 25 Fig. P.1.6.7 EF stat methods & Te Standard Distributions by) Forx=21,2 ~2.25 Hence P(x > 21) = P(2>-2.25)=P(Z<2.25) = 1 - P(Z > 2.25) = 1 - 0.0122 = 0.9878 Fig. P.1.6.7(a) GO-30) «9 and for x = 35, 2= 3) co) Forx=30,2= Hence P00)-P(Z> 1.25) = 0.5-0.1056 = 0,304 1.25 Fig. P.1.6.7(b) 6.87 A radar unit is used to measure speeds of cars on a motorway. The speeds are normally distributed with a mean of 90 knvhr and a standard deviation of 10 km/hr. What is the probability that a car picked at random is travelling at more than 100 km/hr? Soin. : Let x be the random variable that rej to find the probability that x is higher than 100 or P(x > 100) (100-90) _, For x = 100,z="—q9 = presents the speed of cars. x has 1 = 90 and o = 10. We have P(x>90) = P(z>1)=0.1587 The probability that a car selected at a random has a speed greater than 100 km/hr is equal to 0.1587. 1 Fig. P.1.6.8 the length of time between charges of the battery is normally and a standard deviation of 15 hours. Mr. Sharma owns bability that the length of time will be For a certain type of computers, distributed with a mean of 50 hours ‘one of these computers and wants to know the pro! between 50 and 70 hours. standard Distribution, 1 of $0 and a standay of time. It has a mea Soln.: Let x be the random variable that represents the length deviation of 15. We have to find the probability that x tetween 50 and 70 oF PC 50n-P22 130° 0.5 - 0.0918 = 0.4082 ‘The probability that Mr. Sharma's computer has a Jength of time between charges of battery and 70 hours is equal 00.4082 139 Fig. P.1.6.9 716.10: The length of similar compononts J eg ot eS a by a company are approximated by @ normal soa a Tanda ‘a standard deviation of 0.02 cm. If @ component a) Whatis the probabil soon? probability that the length of this ‘component is between 4.98 and b) Whats the probabi _ com 5,04 om? ty thatthe length ofthis component is between 4.96 and (a) (4.§ ) P98 <5.02) = P(-1ez DI = 1-(P(Z>1)+PZ> 1)] due to symmetry = 1-2 0.1587= 0.6826 1 fi Fig. P.1.6.10 BE stat. Methods & Test of Hypo.(B.Se.-Comp-MU) 1-53 Standard Distributions eT EEE (b) P(4.96a,) = 0.15 a-2 a, = 1.04, aS = 104 a = 724§x1.04=77.2 from which X = 77.2. The top 15% of the students will have an average 77.2 or more and will receive an A. Ex. Let Z ~ N[O, 1]. Find the value of a such that P(Z < a) = 0.01 (i.e. find the lower 1% point). Soin. P(Z-a) = 0.01 a= -233 Ex 1.6.13 : The germination success rate for begonia seeds is 60%. In a package of 200 seeds, what the probability that over haif of them germinate, i.e., what is P(X >100)? ‘Solin. : X : Number of begonia seeds germinated X ~B (200, 0.6) ‘As nis large distribution of X is approximated to normal with mean = np and variance = npq Mean = Variance = X ~N(120, 48 np = 200 x 0.6 = 120 npq = 48. = 120° P(X> 100) = P(2> ta om ) = P(Z>-2.89) P(Z<289) due to symmetry 1-P(Z>2.89) 1- 0.0019 = 0.9981 Pe 16.18: Soin. : Let X : number of sales. ‘A telephone-soliciting company has a 12% success rate for sales. If the callers contact 500 People, find the probability of getting at most 64 sales. X ~B (500, 0.12) As nis large distribution of X is approximated to normal with mean = np and variance = npq Mean Variance X ~N(60, 7.27) P(X <64) = P(z> np = 500 x 0.12 = 60 npq = 52.8. 64— Ta7 )=P(Z>0.55) 6 Ex 1.6.15 : Of large group of person 5 per cent are under 60 inches in height and 40 per cent are between 60 and 65 inches. Assuming Normal distribution, find mean and standard deviation of height. Soin. : X: uy Height of persons. X~Niu, 9) P(K < 60) PZ -a) a 0.05 and P(60 < X <65)=0.40 P(2>S=# = 0.05 0005 wherea= =" aco 0.05 - 1.64 =>-a= 164 (@ stat. Methods & Test. of Hypo. Standard Distributions > os = - 164 e(L) P(OO Pla 1-[P(Zb)] = > P(Z>-a)+P(Z>b) = 1-04=06 > 0.05+P(Z<-a)+P(Z>b) = 06 > 0.05+P(Z>b) = 06 > P(Z>b) = 055 bis negative > PiZ P(Z>-b) = 0.45 = -b = 013 b = -013 > Soh = -03 > 65 = p-0130 2) Solving (1) and (2) simultaneously 60 = p-1.640 2233112 w= 65 40.130 = 65 + 0.13 x 3.3112 = 65.4305 a i = ai X ~ N(W=65.43,0=3.31) Syllabus Topic : The Chi-square Distribution 1.7 The Chi-square Distribution 1. What is chi square distribution ? 12. Define chi square distribution. The chi-squared (x°) 0) is given by 1 fix) = Tp 1 ort) For x $0, fix)=0 . This density is written in terms of the gamma function. Some of the key properties of ths function are | © Myee-Mx- Ds 1 . 3) =VR: ©) Mix)=(x- 1)! ifxis a natural number, “The degrees of freedom n, define the shape of the 7density. For n <3, the density has a mode a zero. For n 2 3, the mode moves further away from zero as n increases. The shapes of some specific densities are given below. Graph of sever’ ch-equared denaies Fang P Fig 1.7.1 : Graph of several chi square densities Use of 77 tables for calculation of probabilities Let Y be a’ varis % variate with nd. f. Let o be some probabil . : ty between 0 percentile of» chi-square disbuion with n depres of feedoms the cane geen ee at wader the curve and to the right of 7). is on PY2x) =a Example 1 If X~ 7, then find PIX > 0.584) Now, from 7/ table PIX > 0.584] =0.9 ee PEt sua me eS Example 2 Suppe Now, Example 3 Suppc —_— — DTA Pre iQ. State 1 XE 2m 2 XI and \ mean 30 OFX 2 a= varia 4 EX respe 5 Let z one J oN Let devis squa paras Methods & Test. of B.Sc.- Comp.-MU) Standard Distributions Fig. 1.7.2 Example 2 ‘Suppose X ~ x Find P(X < 14.067). Now, from tables we can find, P(X >14.067) = 0.05 => P(X < 14.067) = 1 - 0.05 = 0.95. Example 3 Suppose X ~ ie Find P(3.940 < X < 18.307) P(3.940 < X < 18.307) = P(X > 3.940) - P(X > 18.307) = 0.95-0.50=0.45 Syllabus Topic : Properties of Chi Square Distribution 7.1 Properties of Chi Square Distribution fe Pp State Properties of Chi square distribution (x°). If X has a gamma distribution with paramters m and B then i. has a chi-square distribution with 2m degrees of freedom. TEX has a chi-square distribution with n degrees of freedom, then the mean of X is 1, = E(X) =n and variance of X is 2na ur X has a chi-square distribution with n degrees of freedom, then the mean of Y is pty = E(Y) = n6. If X has a chi-square distribution with n degrees of freedom, then the variance of X is @, = B(X ~ p,)?) = 2n. f has a chi-square distribution with n degrees of freedom, then the variance of Yis 6, = 2nc", IE X and Y are independent chi-square random variables with n and p degrees of freedom respectively, then X + Y isa chi-square random variable with n + p degrees of freedoms, Let 2, .., 5 be independent normal random variables with mean zero and standard deviation one and let $= Z; + +2, . Then S has a chi square distribution with n degrees of feeders Let Uy, WU, be independent normal random variables with the same mean and standard = U4 +U, ~ deviation 6 and let G = #1 sou w = (y, - ye Square distribution with nm - 1 degrees of freedom an Parameters © ang 20°. (u,-G)* Then W has a chi id W is a gamma random variable with jstribution with © def i distil RRS) BeP nase ze oy Ye — 0 x? £74.39 6X1, XX se suns eo ites [> ° xX, ] 1 Xa ‘As X;, Xx Xa Xeo eid. N (0,1) variates then uae at rex, - hy 0.1=09 2 of cam] =} Pl.” 65.227 to 0 ofa eosa2|= Pf 1 Syllabus Topic : The {distribution . 8 __The Student t-distribution Define t-distribution. Introduction fe have two independent random variables Y and Z, such that: Definition : Suppose that w 2 y ~N(0, 1and Z~%, ‘Then the random variable X defined by tas a distribution with n degrees of freedom- denoted y- Probability density function ntl +) eel nm Eyl Fae The t:distribution is symmetric about zero and its general shape is like the bell-shape of anormal | ee the tails ofthe -distibotion can approach zero much more slowly than those of nor distribution- ie. the t-distribution is heavier tailed than the mal. | define how heavy-tailed the t-distributio ae Note: The t- distribution with n = 1 is sometimes referred to as the Cauchy distribution. This is 5° auc ted ts reen aed veronce do vick exit | (Tia ls beomuse the integral pecitying the mean and variance are not absolutely convergent) 7 | tat. Methods & Test. of Hypo. Comp.-MU) ‘Standard Distributions Important note ‘The density of a t-distribution converges to that of the standard normal as n — ©. ‘The diagram below shows how the t-distribution varies for different degrees of freedom. ‘Comparing several t-distributions with the standard normal. Comparing soveracstibuons wth the standard normal : = z = ie ' Fig. 1.8.1 1.8.1 Probabilities Probabilities associated with the t-distribution can be looked up in tables. The probability P[It,|>c] = 0: for various values of n and c ( for two tailed) P[1,2>d] = @ for various values of n and d ( for one tailed) Example 1 Let X ~ ts. Then P(X > 2.352) = 0.05 (As one tailed). using t-able 2Talled 0.40 0.90 020 015 0.10 {Tailed 0.20 015 0.10 0.075 0.05 at 1 2 3———_______> 2.182 Fig. 1.8.2 Example 2 Let X ~ty,. Find P(X < 2.179). P(X < 2.179) = 1 - P(X > 2.179) = 1 - 0.025 = 0.975. 1-60 Standard FP sae Methods & Test. of Hypo. Exam Find the 5% point for ts. Directly from tables, this is seen to be ty, 99s= 1.943. (Thus P(X > 1.943) = 0.05) Example 4 Let X ~ tyo, Find the value of ¢ such that P(X 2.764) = 0.01 => P(X <— 2.764) = 0.01. So, the lower 1% point, ¢, is - 2.764. Example 5 7 Let X ~ tis. Find i) P(X > 1.517) ii) P(X < 2.603) ii) PUXI> 1.753) iv) POX! < 2.131) VtifP(X>)=0.1 vi)uif P(X 1517)=0.075 Gi) PK < 2.603) = 1 - P(X > 2.603) = 1 - 0.01 = 0.99 Gil) P(X1> 1.753) = 0.10 (corresponds to two tailed) (iv) POX < 2.131) = 1- PCIXI> 2.131) = 1 - 0.05 =0.95 @) PK >) =0.1 t= 1341 (i) PK P(X >-u) = 0.2 = 0.8666 u= - 0.8666 Note: To find non-standard percentage points (such as the 12.5% point, for example), we need to = the fdistribution function table, Syllabus Topic : Properties of t-Distribution ~1.8.2 Properties of t-Distribution |. _State properties of t-distribution. ‘The t-Distribution Characteristics of the t-distribution similar to the normal distribution. Itis bell-shaped. It is symmetric about the mean. The mean, median, and mode are equal to 0 and are located at the center of the distribution. The curve never touches the x axis. The t-Distribution Characteristics of the t-distribution that differ from the normal distribution The variance is greater than 1 2. The t-distribution is a family of curves based on the concept of degrees of freedom, which is related to sample size. fon - 3. As the sample size increases, the t- distribution approaches the standard normal distribution. BF sat Methods & Test. of Hypo.(B.Se.-Comp-MU) 1-61 Standard Distributions 5 SS Se SOE Syllabus Topic : The F-Distribution 19 _The (Fisher's) F-Distribution '@. Define F-distribution. Definition : Consider two independent random variables Y and Z such that nY~ x, and mZ~y_ ‘The random variable X defined by y x=F is then said to have an F-distribution with n and m degrees of freedom- denoted F,, es") a) RES = a2 fhe) BP fF) = ) 2) @ F [i + af f>0 2a The F-distribution therefore has two parameters, both of which are degrees of freedom. The order of the degrees of freedom is important! The F, , distribution is not the same as the F,, distribution. | Note: The density for the F-distribution is only defined for positive values of X. The values of the two degrees of freedom define the shape of the distribution. Plots of the F-distribution for various values of n and m are shown below. Graph of sovera Fcistbution Fig. 1.9.1 Lindley and Scou do not have tables for looking up probabilities associated with the F distribution, FT sian Methods & Test. of Hypo.(B.Sc.- Comp.-MU) _1-62 Percentage points or Separate tables giving 10, 5, 2.5, 1, 05 and 0.1 percentage points for F-distibutions different combinations of degrees of freedom can be found in Lindley and Scott. We will denote the (upper) 100a % point for the F,, ,distribution by F,mq- If X ~F, ., then: PIX >F, ma) = 0 In the table of the 100ct percentage points for the F-distribution, the first degrees of freedom i denoted v, and listed along the columns. The second degrees of freedom is denoted by v, and listey down the rows. Extract : 1% points of the F-distribution Using F-table of 5% or 1% v= 1\2/)3/)4 : | 2 3, Tamed 28.24 The (upper) 1% point for an F, , distribution is 28.24. We write F301 Example Find the 5% point for both the F,,.) and the Fg. distributions. From the 5% points table: Fs g995 = 3.326 Fiosoos = 4.735 Notice that these are not the same. Result ‘Suppose that X = x. F,,,. Then X= Z. Fan Proof y X = Z~F,mifnY~x, and mZ~x.. But by definition of the F-distribution, this means that Zz ¥~F.0€8 required. ‘We can use this result to find lower percentage points for F-distributions : Example 1 : Let X ~ F,19. Suppose we wish to find x such that P(X < x) = 0.05 i.e, we want to find the lower 5% point of the F, 19 distribution. The lower 5% point of the Fs,9 distribution is the reciprocal of the upper 5% point of Fis distribution. (DT sua thods & Test. of Hypo.(B.Sc.- Comp-MU) _ 1-63 ‘Standard Distributions So, 1 = a735 = 02112 iasoos Example 2 : Suppose X ~ F,,. Find the upper and lower 10% points. ‘The upper 10% point can be found directly from tables : Fao. = 2.961 The lower 10% point is the reciprocal of the upper 10% point of the F,. distribution : 1 r979 = 0.2513 Lower 10% point = Fyjq9= Fao Syllabus Topic : Properties of F-distribution .1.9.1 Properties of F Distribution 12, State properties of F distribution. ‘© The F-distribution is positively skewed and with the increase in the degrees of freedom n and m, its skewness decreases. © The value of the F-distribution is always positive, or zero since the variances are the square of the deviations and hence cannot assume negative values. Its value lies between 0 and 0, © The statistic used to calculate the value of mean and variance m Mean = for m>2 2m (n+m-2) (m= 2)" (m4) ‘© The shape of the F-distribution depends on its parameters n and m degrees of freedom.) Variance = . for n> 4 is: ‘© The values of the area lying on the left-hand side of the distribution can be found out by taking the reciprocal of F values corresponding to the right-hand side and the degrees of freedom in the numerator and the denominator are interchanged. This is called as Reciprocal Property of F-distribution, Symbolically, it can be represented as 1 Fymn F.-anm = ‘This property is mainly used in the situations when the values of the lower tail F values are to be determined corresponding to the upper tail F values. [semen] ——_ Continuous R-V. Saree mando VA riable Dert Random Vareti A vx) = ECX’)- EO) where E(x’) = x xP 5 4, Cumulative Distribution Function Where E(X) = J x" f(x) dx. Fa) = Pecss= & poy 4, Cumulative Distribution Function. | 5. p.m-f. from CDF. i Pix) = FO)-Fx-1) roo = J foney 6 eg. “ 1.No. of children in a family. 5. p.df.from C.D. F. 2. No. of errors in a programme. f(x) = £ F(X) 6. Distribution 1. Time to complete programme. © 2. Weight of oil bag. p.n.iip.dt ~ Sr. Distribution Ni Characteristics 1, | Binomial ‘ (Discrete) (1) Used when outcome is Bernoulli Q) Mean=np Variance = npq @) Additive Property If X ~ Bon, P) and Y ~ (ny P) |) - | independent then X + ¥ ~ B (n, +9) er —™ [FF stat. Methods & Test. of Hypo.(B.Sc.- Comp.-MU)_1-65 Standard Distributions Sr. | _ Distribution pan. / pdf. Characteristics No. . a 2 Normal 1 -! cay (1) Used for normal situation like ° fxye—=e 20) xerR i (Continuous) - | oan” height, 1Q = = pe variance = Denoted by (2) Mean=y vari ; X-NL.o?) (3) Curve is bell shaped, symmetric . about jt (4) Additive property if 2 X=N(H 9) X and Y-N (Hy o,) are independent 22 X+Y-N(i, +H 6,+9,) (5) IfX ~B(n, p) and n is large then X ~N (np, npq) 3, | Chi-Square 1 3H wn (1) Mean = _ Variance = 2n Distribution {= rae © (2) IX ~72 nis large there Continuous D0 Denoted by X ~ x, 2 GB) UX-x% 2 Yo hep X and Y are independent. 2 thenX+Y~2X,, 4, | t= Distribution rapt 2 =at1 | (1) Curve is symmetric about zero (continuous) {0=—>= (1 +) bell. ang (2) Mean = Median = Mode = o> (3) Asn increases t-Distribution 5. | F Distribution (1) Distribution position skewed. Continuous m @) Mean = 5 2m’ (n+ m-2 Variance = 2 (2+m=2). ariance = ym — 2)" (m—4) | Fae =F _| sma 10. Discrete Random Variable Ccasify the following random variables 38 discrete or continvous: ‘The number of automobile fatalities Ove 2 72-hour period ‘The length of time to rum a 10-yard dash ‘The diameter of pebbles from a steam switchboard at a city hospital The number of incoming phone calls 10 4 specific location ‘The barometric pressure at noon at 2 ‘The numberof bills paid on time per month 10 & plumber Define random variable with example Define discrete random variable with example Define Continuous random variable with example Define p.m.f, c.£, mean and variance of discrete random variable. ‘Among ten applicants for an open postion, six are women and four are met Three applicants are randomly selected from the applicant pool for final interviews. Let X be the number of female applicants among the final three. Find probability distribution of X. ‘and let the random variable X denote the sum of 2 3 a 5 6 Suppose that a pair of fair dice are to be tossed, the points. Obtain the probability distribution for X. 12 [ [ eee 2) [10 i x {2 [3 [4 [5 [6 Pox =x) | 1/36 | 236 | 3/36 | 4/36 | 5/36 | 6/36 | 5/36 4736 | 3/36 2/36 | 1/36 Let X be a random variable giving the number of aces in a random draw of 4 cards from an Construct a table showing the probability distribution of X. ordinary deck of 52 cards. = e 1 2 3 4 P(X =x) | 0.7187 | 0.2555 | 0.0249 | 0.0007 | 0.0000 Determine ¢ such thatthe following functions are p.m.f.s @) P(K=x)eex; x=1,23,4....10 (i) PM =x)=er x=1,23 Verify whether the the following can be looked upon as p.m.f. for the given values of X. ; alues of X. 1 @ PX=y=4;x=012 (Ans, : 1/55,1/14) AT sa, Mata ot ‘Se.- Comp -MU) Standard Distribution: x41 (i) POR =X) xe 01.23 (iii) Pen =A, x=1,2,34 (Ans, : No, Yes, No) 11, The number of light switch turnons at which the first failure occurs is a r.v. X who sep. m.f. is given by : Pa) = c(t) ‘aet23 Find ) ¢ ii) Calculate the probability that the first failure will not occur until the 10th turn-on iii) Determine the corresponding c.d.f. (Ans, : 1/10,0.349) 12, Arandom variable X has the di 1 3/5 16 | 16 a Find the expected value, variance, and standard deviation of X. (Ans. : 2.9375, 5.5585, 2.3576) 13. A discrete random variable has following probability distribution 4}5 |6|7 3c | Se | Te | 9c | Ie (Find (ii) P< X <6) (iii) Distribution function of X_ (iv) Mean of X 14. A hotel has two conference rooms and on any given day, the number of conference rooms that will be rented out for events is a random variable X. Let X has the following distribution x jo fi [2 PK) | 10} 7 | 8 25 | 25 | 25 (a) Find B(X) and Var(X). (b) Suppose each conference room rents for 1000 a day. Let Y stands for the rents collected cn any given day. Write down the probability distribution of Y. (©) Find B(Y) and Var(¥). (Ans. By 1S. If probability that the value of certain stock will remain same is 0.44, the probabilities that its value will increase by 5 Rs. or Rs.10 per share are respectively 0.19 and 0.23 and the probability that its value will decrease by Rs. 2.50 per share is 0.14. What is expected gain per share’? (Ans. : Rs. 2.90) Stat. Methods & Test. of Hypo. Comp.-MU)_1-68 Standard Ditto, Naining 16. Consider an experiment of drawing 2 balls at random without replacement from a jar contaj 2 black balls and 3 white balls. Let X be the number of black balls selected. Find probatijs, distribution of X, C.D.F. of X 17. Suppose a random variable X has the following probability mass function, x 1f2]/3]/4]5}]6]7]8 P(X =x) | 0.11 | 0.07 | 0.13 | 0.28 | 0.18 | 0.05 | 0.12 | 0.06 Find C.D-F of X, mean of X, variance of X and P(X = 4/X > 2) 18. Let X be discrete random variable have following distribution function [x SA et iLO dd Ibe i eek) FQ | 0.025 | 0.12 | 0.28 | 0.44 | 0.59 | 0.76 | 0.92 | 1 Find p.m.f. of X, P(X > 0/ X > ~ 1), E(X) | Continuous random variable 1. Define Continuous random variable, p.d.f., cumulative distribution function of X 2. What is distribution function of continuous random variable? State properties of D.F. of X 3. Define mean , variance and reliability function of continuous random variable. Write significance of reliability function. 4. It has been determined that the probability density function for the wait in line at a counter is | given by, { 0 ift0 =0 otherwise 7. Ifthe p.d.f. of a random variable X is f(x) = kx for0 43.8] (Ans. : 0.05) Lax =X), then Find i) P(X > 25) il) PX<27.5) ii) P(30.6 << X < 32.8) (Ans. :0.05,0.975,0.005) Distribution 3 Define t- distribution. Write main characteristics oft distribution. Let X ~ t; then find F (X > 1.356), P(X < 1.782), P( tl > 2.179), P(t < 2.681) (Ans. : 0.1, 0.95, 0.05, 0.98) Find the 5% point for ty (Ans. : 1.746) CET sat Methods & Tost. of Hyp.(8.Se~Comp-MU)_ 1-72 Standard Distnion, 4. Find «such that P(X <1) = 0.1 where X (Ans. :~133) F-Distribution 1. Define F- distribution. Write main characteristics bf F-distibution. 2. State reciprocal property of F-dstribution 3. Find the 5% point for both the Fg and the Fo, distributions. (Ans. :3.071733472) 4. LetX~Fygap Find X such that P(X < x) = 0.05 (2.3879) 5. FFind upper and lower 1% of Fs, Qa 2A __Introduction —————_ @. _Explain Concept Hypoth * Hypothesis testing (or Probability, used to test need to test claims abou a) Average height of >) A cigarette man cigarene. ©) Proportion of une To test above claims it ‘on sampling and check called testing of hypoth In testing of hypothesis are called parameter. Fo If Population parameter and which is known 3 Proportion etc. ‘22 _ Statistical Hypo ° Definition : A statistica Which may or may not . According to Prof. Mor about a population”. ~ Hypotheses are two typ « — / Nutt Hypothesis , 6 The we.

You might also like