Scribd
Upload
32 views8 pages

Law of Large Number

Law of large number.

Uploaded by

mrkambakth1998
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
32 views8 pages

Law of Large Number

Law of large number.

Uploaded by

mrkambakth1998
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
You are on page 1/ 8
—_—_ erating FUNCTIONS AND LAW OF LARGE NUMBERS : 7:95 experiment is t0 f088 two balls into fi 13 - to five boxes in suc hat is the pad. of X 7 (if) Giv first box. Wh ha (i) Give the mgs. of X. (iii) Find th a. show that the r® cumulant for the distribution : le mean and vance of aye ce? Gand OS x 0; 4 nd the characteristic function of ity? ps 7 (: s *) 16 (0) Find the characteristic function of the following distributions and their variances 1 Alyy (20 0, x? 0), (i) dF) = Gi) PK == "GP qh, (O0 few )\% 0, _ elsewhere, show that the characteristic aeton aiken icieiee t the Perera rot x and Y. Commer the result. t,) is the join 78, (a) Let K(ty 2) = log M(ty 2) where M(ty 2, 2, FKOD « vark; FKOD) ~ Cov (X,Y) ft at, Ot sadom variable with the possible values reed with te joint probabilities given product of the characteristic t mgf, of X and Y. Show that : aK(0,0) . E(X); a () % Nisa two-dimensional Son oe x, and also and 10" Y- in the adjoining Teble- Find the characteristic functions vi(h). walt) Pa ex, ) respectiviely and show that (ty 2) = sath walla) When Poo Pa = Por P10 oD Also interpret (")- 74, (a) A random variable X has 2 an value of Sand variance of 3. (j) What is the least value of ix-51 <3)? (i) What value of hk guarantees th | 2) °) and compare it with the upper bound obtained by Chebychev's inequality. ans. Upper bound by Chebychev’s inequality = 0-444; Actual probabilty = 0-134 gan. (8) HEX Xpy = Xq be rn with means jy, fy, Hy and standard deviations 64, 69). Oy respectively and [Var(X +X X,)] 0? > 0 as noe show that (Sx) converge to zero in probability, Hence show that if m is the number of successes in n independent trials, the probability of success at the ith trial being p, then m/n converges in probability to (p, + p+ + P,)in. (b) IX, takes the values 1 and 0 with corresponding probabilities p, and 1 — p,, examine whether the weak law of large numbers can be applied to the sequence {X,] where the variables X,, = 1,2,3, .. are independent. 1 Hint. Var{X,) = Pn SG? yes, W.L-L.N. holds good. 743. (a) [Xj], k = 1, 2, .. is a sequence of independent random variables each taking the values =1, 0, 1. Given that P(X, = 1) = i PUX, = = 1), P(X, = 0) = 1-2, examine if the law of large numbers holds for this sequence. ‘Ans. WLLN holds. (6) Examine whether the weak law of large numbers holds good for the sequence X,, of independent random variables, where of. vn ‘Ans. WLLN holds. 7A. (a) (X,) is a sequence of independent random variables such that 1 e-h)=m omeeh)ion Examine whether the weak law of large numbers is applicable to the sequence (X,_)- Hint. Var(X,) = Py S 3. Yes. (#) IX is a ro, and E(X2) < a, then prove that P{|x| > a} s (x2), for all a>0. « Use Chebychev’s inequality to show that for n> 36, the probability that in n throws of a fair die 31 the numberof sixes les between n= Vi and 2+ is atleast 3+ Hir Int Chebychevs Inequality gives : ‘3 1 Since X20, Z-Ji>0 > 1236. i wy (i) PX> ne) () Obtain the generating func . 4 aera eds cee tM Of ty the probability that in m tosses of an ideal coin, no run (©) Let X be a non-negative integral-valued a indom variable with, Probability generating function (@) probability generating. f ‘netion of X given that Y = x 3 ig FUNCTIONS AND LAW OF LARGE NUMBERS pneRatiN 7:99 spa sequence of Bernoulli trials, let Uy be the probability that the first combination SF occurs 12. rials number (n - 1) and n. Find the probability Generating function, mean and variance, | Mean=".; Variance = pint PIS) 7 em. Ina sequence of Bernoulli trials, let U, be the probabilty of an even number of successes. Prove the recursion formula U, = qU ja + (1 ~ U,.4) p. From this derive the probability generating function and hence the explicit formula for U, a Sy mle tact ating Function : G(t) = )°u,t" =) Sean ans. Generating Function R [2 }« Oe 4." Talat zL1+(1-2P)"] 723. A series: of independent Bernoulli trials with probability ‘p’ of success for each trial is performed until an uninterrupted run of r successes is obtained for the first time where r is a given positive integer. Assuming that the probability of a success in any trial is p = 1 - q, show that the pis’ (1-ps) 1—s+qp'stt 724. A population of N distinct elements in equal proportion is sampled with replacement. If S, denotes the sample size necessary for the acquisition of r distinct varieties, and p,(r) = P(S, = 1), prove the recursion formula probability generating function of the number of trials is: F(S) = N-r+1 Peal?) Pal?) + P,(t-1) Hence, or otherwise, derive the generating function P,(s) of pp). | (N=1)(N = 2) .(N=141) | ns E PyS) (N-s)(N 2s)... {N -(r=1)s} 725, There is a complete set of values (events) xq Xy Xp Xp « With respective probabilities py py, Py = jy ~» Which vary with time. These are related by the differential equation : a id= *(Pi-a Pa) Find the value of py Hint Post multiplying (*) with s* and summing over k from 0 to =, we get A fe] = M5609 -Gi9]=*(6-1)(9) Integrate both sides w.r.t. ‘t’ and use G(1) = 1 to determine the constant of integration. Mak Kt 26 In a sequence of Bernoulli trials, let A, be the event that a run of m consecutive successes Poze Fem 1 Sccurs between the 2"th and 2"*7th trials. Show that if P2>, there is probability one that infinitely many A, occur; if psy, then with probability one only finitely many A,, occur. 1 * For a given sequence (X,) of 1.2, ‘lt Xp) = (sin nt) / nt. a 7-100 7:28, 7:30. 731. 732. 733, 734. 738. 736. FUNDAMENTALS OF MATHEMATICAL STN ence show that even though the sequ termi distribution function of X,,. Hence s es, ce rmine distribu 3 avenge toa limit g(t), the sequence of distribution functions at converge Wo 0 claritin meio Wii! the’ colton that ts violated here > WY, = “4-3, is known and finite show that it is possible to improve Chebychevs inequity CH in the form Ty+2 P| |X-p|2ko |] << ——*——__, [Ix-nl2ko] Yd aay tnoers X random variable with mean yr and variance o, and kis any postive integer me than 1, (Xp m2 1) is a sequence of random variables such that : pi fi- VanX) = 15 4= 1, 2,3, and cov(X, X) = a otletise: Show that W.LLN. holds for the sequence (X,. Let {X,,=1,2..} be a Sequence of standardised variates and Cor (Xn Xy)=exp{-|mn—nfaa>0 and m + m. Show that WALL. holds for this sequence Two continuous variates X and Y have a joint pif. with a joint characteristic function (ty, tf £343) isthe marginal density, show that 3), the rth simple moment for the conditional distribution of ¥ given X = 2, satisfies the equation : 1 HA. 0) Wa) 900) = 2 vas Prove that the real Part of a characteristic functis that if ya(t) = ay(t) + iby(t) and y(t) = a , (0 ext) ~ (0) bolt) is a characteristic function? haw Bul Tac sry tron cf = 2 lara fre and hence deduce P{\X-E(X)|>ko} < = where k > 0 and Var(X) « 2, Let X be a rv. with moment generating function M(),-h X~P(A+) Py) = exp[(u+t)(s-1)] + Y~P(u+0) Pyolt)= Sp, = exp[-a-p-b+ (+ u)teoe i where p, = P (X + Y =k). P(X + Y= 0) = po = constant term in (*) = expl - A - w - B]. Define convergence in probability. Let X;, Xp, Xy .. be iid. variates with : fayee®-V x21 Show that ¥,, -> 1 in probability, where Y, = Min(X;); 1< k 0, P[]¥,-1] <8) +125 240, a

You might also like