Scribd
Upload
11 views4 pages

Proof Central Limit Theorem

Proof of central limit theorem

Uploaded by

mohammed.rebiai
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
11 views4 pages

Proof Central Limit Theorem

Proof of central limit theorem

Uploaded by

mohammed.rebiai
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/ 4
Proof the Central Lomt Theorem and Related Back ground Trtacmetion on Moment Generating Functions Reference’ / Rice, TA Mathematical Sfetsties and Date Analysis Uh Dux bary Press, Belment, CA (12952, The Moment-Generating Function ‘This section develops and applies some of the properties of the moment-generating function. It tums out, despite its unlikely appearance, to be a very ‘useful tool that can dramatically simplify certain calculations. The moment generating function (mgf) of a random variable X is M()= E(¢*) if the expectation is defined. In the discrete case, MO = PO and in the continuous case, Mw) = f f(x) de “The expectation, and hence the moment-generating function, may of may not exs for any particular value of In the continuous case the existence of the expects 4.5 The Moment-Generating Function \X} depends on how rapidly the tails of the density decrease; for example, since the tails, of the Cauchy density die down at the rate x~?, the expectation does not exist for any ¢ and the moment-generating function is undefined. Since the tails of the normal density die down at the rate e~", the integral converges for al . PROPERTY A If the moment-generating function exists for ¢ in an open interval containing zero, it uniquely determines the probability distribution. = ‘We cannot prove this important property here—its proof depends on properties of the Laplace transform. Note that Property A says that if two random variables have the same mgf in an open interval containing zero, they have the same distribution For some problems, we can find the mgf and then deduce the unique probability distribution corresponding to it. ‘The rth moment of a random variable is E(X’) if the expectation exists. We have already encountered the first and second moments earlier in this chapter, that is, E(X) and E(X*), Central moments rather than ordinary moments are often used: The rth central moment is E{{X — E(0))’}. The variance is the second central moment and is a measure of dispersion about the mean. The third central moment, called the skewness, is used as a measure of the asymmetry of a density or a frequency function about its mean; if a density is symmetric about its mean, the skewness is zero (see Problem 72 at the end of this chapter). As its name implies, the moment- generating function has something to do with moments. To see this, consider the continuous case: MO = f $0) de ‘The derivative of M(t) is mM ap > e* f(x) de $f “ee In ean be show that eitferentation and iteration canbe interchanged, so that M()= f xe"f(x) de and M0) f af (x) dx = E(X) Differentiating r times, we find MO) =E) It can further be argued that if the moment-generating function exists in an interval containing zero, then so do all the moments. We thus have the following property. Proof 5.3 Comergence in Distribution andthe Cental Limit Theorem 18 ‘The central limit theorem is concerned not with the fact that the ratio S,/n con- verges to yx but with how it fluctuates around jz. To analyze these fluctuations, we standardize: S,- mu n= OT It is easily verified that Z, has mean 0 and variance 1. The central limit theorem states that the distribution Of Z,, converges to the standard normal distribution. THEOREM B_ Central Limit Theorem Let X,,X,... be a sequence of independent random variables having mean 0 and variance o? and the common distribution function F and moment-generating func- tion M defined in a neighborhood of zero. Let 520% Jin? (Sigss) 00, -w 0. Since E(X) = 0, M'(0) = 0, and M"(0) = 07. As n> 00, t/(0 Vn) —> 0, and (si) =144 (sim) +e ayn} 27 on, where ¢,/(F /(no”)) + as n + 00. We thus have ; Itcan be shown that if @, —> a, then and « li 1 im, (14 From this result, it follows that Mz (0 2 asn—> 0 EXAMPLE C EXAMPLE D M0 Chapter 5: Limit Theorems where exp(/?/2) is the mgf of the standard normal distribution, as was to be shown, # ‘Theorem B is one of the simplest versions of the central limit theorem; there are many central limit theorems of various degrees of abstraction and generality. We have proved Theorem B under the assumption that the moment-generating functions exist, which is a rather strong assumption. By using characteristic functions instead, wwe could modify the proof so that it would only be necessary that first and sec~ ond moments exist. Further generalizations weaken the assumption that the X; have the same distribution and apply to linear combinations of independent random vari- ables. Central limit theorems have also been proved that weaken the independence assumption and allow the X; to be dependent but not “too” dependent. Central limit theorems are still an active area of research in probability theory. For practical purposes, especially for statistics, the limiting result in itself is not of primary interest. Statisticians are more interested in its use as an approximation with finite values of n. It is impossible to give a concise and definitive statement ‘of how good the approximation is, but some general guidelines are available, and ‘examining special cases can give insight. How fast the approximation becomes good depends on the distribution of the summands, the X;. If the distribution is fairly symmetric and has tails that die off rapidly, the approximation becomes good for relatively small values of n, If the distribution is very skewed or if the tails die down very slowly, a larger value of n is needed for a good approximation, The following examples deal with two special cases. ‘Since the uniform distribution on (0, 1] has mean 4 and variance qb, the sum of 12 uniform random variables, minus 6, has mean 0 and variance 1. The distribution of this sum is quite close to normal; in fact, before better algorithms were devel- oped, it was commonly used in computers for generating normal random variables from uniform ones. It is possible to compare the real and approximate distributions analytically, but we will content ourselves with a simple demonstration. Figure 5.1 shows a histogram of 1000 such sums with a superimposed normal density function. The fit is surprisingly good, especially considering that 12 is not usually regarded as alarge value of. # ‘The sum of nm independent exponential random variables with parameter A = 1 fol- lows a gamma distribution with = 1 and a =n (Example F in Section 4.5). The ‘exponential density is quite skewed; therefore, « good approximation of a standard- ized gamma by a standardized normal would not be expected for small n. Figure 5.2 shows the caf’s of the standard normal and standardized gamma distributions for increasing values of n, Note how the approximation improves as n increases. *

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy