The summations are over ω: µ(ω)= m. However, <strong>the</strong> number <strong>of</strong> suchsequences <strong>of</strong> ω’s that µ(ω) = m is clearly equal to <strong>the</strong> number <strong>of</strong>mpossible selections <strong>of</strong> m symbols out <strong>of</strong> n, C . And so,P{ µ = m} = C m np m q n-m (3.1)which is <strong>the</strong> ma<strong>in</strong> formula <strong>of</strong> <strong>the</strong> Bernoulli trials.Its <strong>the</strong>ory is seen to be almost trivial but not trivial is to learn how toapply it, that is, how to f<strong>in</strong>d those phenomena that are sufficiently welldescribed by that pattern. A classical example <strong>of</strong> <strong>the</strong> trials is a toss <strong>of</strong> aco<strong>in</strong>, but when attempt<strong>in</strong>g to discover someth<strong>in</strong>g more <strong>in</strong>terest<strong>in</strong>g, weenter <strong>the</strong> doma<strong>in</strong> <strong>of</strong> doubtfulness. Thus, is it possible to consider abirth <strong>of</strong> an <strong>in</strong>fant <strong>of</strong> one or ano<strong>the</strong>r sex as a Bernoulli trial (<strong>and</strong> regarda male birth, say, as a success)?Accord<strong>in</strong>g to genetic ideas, this is quite natural. However, thoseideas lead just as naturally to <strong>the</strong> frequency <strong>of</strong> male births p = 1/2whereas it somewhat exceeds 1/2 as established by exam<strong>in</strong><strong>in</strong>g such animmense material that it becomes impossible to question it. Then,however, it is perhaps permissible to admit <strong>the</strong> opposite hypo<strong>the</strong>sis <strong>of</strong>p ≠ 1/2? Once more, no, s<strong>in</strong>ce <strong>the</strong> Bernoulli trials presume a constantprobability <strong>of</strong> success whereas <strong>the</strong> statistical data certa<strong>in</strong>ly <strong>in</strong>dicatethat <strong>the</strong> frequency <strong>of</strong> male births <strong>in</strong>creases after long wars. Thedependence <strong>of</strong> <strong>the</strong> probability <strong>of</strong> male births on <strong>the</strong> social conditions<strong>of</strong> <strong>the</strong> family [<strong>and</strong> on o<strong>the</strong>r circumstances] is also be<strong>in</strong>g discussed sothat <strong>the</strong> model <strong>of</strong> Bernoulli trials does not <strong>in</strong> this case completelycorrespond to reality.Then, statistically <strong>in</strong>vestigat<strong>in</strong>g that frequency we f<strong>in</strong>d out that,strictly speak<strong>in</strong>g, <strong>the</strong> model <strong>of</strong> those trials is unacceptable; however,s<strong>in</strong>ce <strong>the</strong> probability <strong>of</strong> male birth is never<strong>the</strong>less very near to 1/2, it isonly possible to reject <strong>the</strong> hypo<strong>the</strong>sis <strong>of</strong> its applicability throughstatistical research based on pr<strong>of</strong>ound corollaries <strong>of</strong> formula (3.1). Wewill see now how it is carried out <strong>in</strong> Chapter 4.An application <strong>of</strong> stochastic methods results <strong>in</strong> a conclusion that,strictly speak<strong>in</strong>g, we ought not to discuss <strong>the</strong> probability <strong>of</strong> male births(or statistical stability). However, <strong>in</strong> <strong>the</strong> f<strong>in</strong>al analysis we will f<strong>in</strong>d outmuch more than had <strong>the</strong>re been an ideal conformity with <strong>the</strong> <strong>the</strong>ory <strong>of</strong>probability: we discover for sure that <strong>the</strong>re exists a still unidentifiedagent regulat<strong>in</strong>g <strong>the</strong> numbers <strong>of</strong> men <strong>and</strong> women.The model <strong>of</strong> Bernoulli trials is <strong>of</strong>ten applied for estimat<strong>in</strong>g someplans <strong>of</strong> acceptance <strong>in</strong>spection <strong>in</strong> which <strong>the</strong> manufactur<strong>in</strong>g <strong>of</strong> faulty(failures) or suitable (successes) articles must be described by thatpattern. However, after recall<strong>in</strong>g <strong>the</strong> discussion <strong>in</strong> Chapter 1 <strong>of</strong> <strong>the</strong>possibility <strong>of</strong> a stochastic description <strong>of</strong> manufactur<strong>in</strong>g faultyproducts, it becomes evident that that model can only be made use <strong>of</strong>when <strong>the</strong> <strong>in</strong>dustrial process is arranged well enough.We will discuss at length <strong>the</strong> attempt to apply <strong>the</strong> same model to <strong>the</strong>problem <strong>of</strong> legal verdicts. Pert<strong>in</strong>ent <strong>in</strong>vestigations are connected with<strong>the</strong> names <strong>of</strong> such first-rate scholars as Laplace <strong>and</strong> Poisson, <strong>and</strong> <strong>the</strong>irstudy is very <strong>in</strong>structive. It shows by an example taken from historythat a perfect comm<strong>and</strong> <strong>of</strong> <strong>the</strong> ma<strong>the</strong>matical methods <strong>of</strong> <strong>the</strong> <strong>the</strong>ory <strong>of</strong>n26
probability can be coupled with an absolutely wrong approach toreality 4 .3.2. Poisson’s jurors. Laplace, <strong>and</strong> <strong>the</strong>n Poisson <strong>in</strong>vestigated <strong>the</strong>issue <strong>of</strong> <strong>the</strong> probabilities <strong>of</strong> mistaken legal verdicts. A certa<strong>in</strong> juror cannaturally make a mistake. Laplace assigned jurors a very modestability <strong>of</strong> correct judgement: he thought that for each separatelyconsidered juror <strong>the</strong> probability <strong>of</strong> a mistake was a r<strong>and</strong>om variableuniformly distributed on segment [0, 1/2]. Poisson did not agree; hera<strong>the</strong>r believed that <strong>the</strong> probability <strong>of</strong> a correct judgement should beestimated by issu<strong>in</strong>g from statistical data. The impossibility <strong>of</strong>precisely establish<strong>in</strong>g whe<strong>the</strong>r rightly or not a given accused personwas found guilty presents here <strong>the</strong> greatest difficulty <strong>of</strong> a directstatistical estimate.Poisson’s ideas widely applied now also consisted <strong>in</strong> that <strong>in</strong> such asituation it was necessary to construct a statistical model with <strong>the</strong>unknown probability enter<strong>in</strong>g it as a parameter <strong>and</strong> to attempt todeterm<strong>in</strong>e it by pert<strong>in</strong>ent data.Let us consider <strong>the</strong> adm<strong>in</strong>istration <strong>of</strong> justice <strong>in</strong> more detail. The trialis based on <strong>the</strong> <strong>in</strong>quest. Denote <strong>the</strong> event consist<strong>in</strong>g <strong>in</strong> that <strong>the</strong>evidence collected at <strong>the</strong> <strong>in</strong>quest was sufficient for <strong>the</strong> trial to declare<strong>the</strong> defendant guilty by A, <strong>and</strong> <strong>the</strong> contrary event by A . Given A, all<strong>the</strong> jurors, provided <strong>the</strong>ir judgement is faultless, ought to unanimouslyvote for <strong>the</strong> prosecution; o<strong>the</strong>rwise (event A ) for <strong>the</strong> defence.Actually, ra<strong>the</strong>r <strong>of</strong>ten <strong>the</strong> votes are divided ow<strong>in</strong>g to mistakes madeby <strong>the</strong> jurors. Poisson’s ma<strong>in</strong> proposition was that such divisionconformed to <strong>the</strong> Bernoulli pattern. If n is <strong>the</strong> number <strong>of</strong> jurors, p, <strong>the</strong>probability <strong>of</strong> a correct judgement <strong>of</strong> each juror, <strong>the</strong> number <strong>of</strong> votesfor <strong>the</strong> prosecution, µ, it is described <strong>in</strong> <strong>the</strong> follow<strong>in</strong>g way.1) Given A, µ is <strong>the</strong> number <strong>of</strong> successes for <strong>the</strong> n pert<strong>in</strong>entBernoulli trials with probability <strong>of</strong> success p.2) Given A , µ is <strong>the</strong> number <strong>of</strong> failures for <strong>the</strong> same pattern.Accord<strong>in</strong>g to <strong>the</strong> French legislation, n = 12 <strong>and</strong> <strong>the</strong> defendant wasdeclared guilty if µ ≥ 7. The probability <strong>of</strong> that outcome isP g = P(A)P{µ ≥ 7/A} + P( A )P{µ ≥ 7/ A } =12 12m m 12 − m m 12−m m12− + −12−m= 7 m=7∑ ∑ (3.2)P( A) C p (1 p) [1 P( A)] C p (1 p) .Crim<strong>in</strong>al statistics provides <strong>the</strong> frequency <strong>of</strong> such verdicts which isapproximately equal to P g <strong>and</strong> Poisson thoroughly checked its stabilityover <strong>the</strong> years. However, expression (3.2) <strong>in</strong>cludes two unknownparameters, P(A) <strong>and</strong> p. Know<strong>in</strong>g only P g , it is impossible to determ<strong>in</strong>e<strong>the</strong>m <strong>and</strong> it is <strong>the</strong>refore necessary to turn to statistics which will<strong>in</strong>dicate not only whe<strong>the</strong>r defendants were found guilty or exonerated,but [<strong>in</strong> one case, see below] by how many votes as well. Thus, be<strong>in</strong>gaccused exactly by seven votes has probabilityP g {µ = 7} = P(A)P{µ = 7/A} + P( A )P{µ = 7/ A } =7 7 5 7 5 7P( A) C p (1 − p) + [1 − P( A)] C p (1 − p) .(3.3)12 1227
- Page 1 and 2: Studies in the History of Statistic
- Page 3 and 4: Introduction by CompilerI am presen
- Page 5 and 6: (Lect. Notes Math., No. 1021, 1983,
- Page 7 and 8: sufficiently securely that a carefu
- Page 9 and 10: is energy?) from chapter 4 of Feynm
- Page 11 and 12: demand to apply transfinite numbers
- Page 13 and 14: for stating that Ω consists of ele
- Page 15 and 16: chances to draw a more suitable apa
- Page 17 and 18: Let the space of elementary events
- Page 19 and 20: 2.3. Independence. When desiring to
- Page 21 and 22: Eξ = ∑ aipi.Our form of definiti
- Page 23 and 24: absolutely precisely if the pertine
- Page 25: where x is any real number. If dens
- Page 29 and 30: Nowadays we are sure that no indepe
- Page 31 and 32: λ = λ(T)with λ(T) being actually
- Page 33 and 34: (1/B n )(m − A n )instead of the
- Page 35 and 36: along with ξ. For example, if ξ i
- Page 37 and 38: µ( − p0) ÷np0 (1 − p0)nhas an
- Page 39 and 40: distribution of the maximal term |s
- Page 41 and 42: ξ (ω) + ... + ξ (ω)n1n{ω :|
- Page 43 and 44: P{max ξ(t) ≥ x} = 0.01, 0 ≤ t
- Page 45 and 46: 1. This example and considerations
- Page 47 and 48: IIV. N. TutubalinTreatment of Obser
- Page 49 and 50: structure of statistical methods, d
- Page 51 and 52: Suppose that we have adopted the pa
- Page 53 and 54: and the variances are inversely pro
- Page 55 and 56: It is interesting therefore to see
- Page 57 and 58: is applied with P(t) being a polyno
- Page 59 and 60: ut some mathematical tricks describ
- Page 61 and 62: It is clear therefore that no speci
- Page 63 and 64: of various groups of machines, and
- Page 65 and 66: nnA(λ) x sin λ t, B(λ) = x cosλ
- Page 67 and 68: of the mathematical model of the Br
- Page 69 and 70: dF(λ) = f (λ) dλ, so that B( t
- Page 71 and 72: usually very little of them. Indeed
- Page 73 and 74: This is the celebrated model of aut
- Page 75 and 76: applications of the theory of stoch
- Page 77 and 78:
achieved by differentiating because
- Page 79 and 80:
u(x 1 , x 2 , t 1 , t 2 ) = v(x 1 ,
- Page 81 and 82:
Reasoning based on common sense and
- Page 83 and 84:
answering that question is extremel
- Page 85 and 86:
IIIV. N. TutubalinThe Boundaries of
- Page 87 and 88:
periodograms. It occurred that work
- Page 89 and 90:
at point x = 1. However, preceding
- Page 91 and 92:
He concludes that since the action
- Page 93 and 94:
The verification of the truth of a
- Page 95 and 96:
In the purely scientific sense this
- Page 97 and 98:
ought to learn at once the simple t
- Page 99 and 100:
the material world science had inde
- Page 101 and 102:
values of (2.1) realized in the n e
- Page 103 and 104:
*several dozen. The totality µ ica
- Page 105 and 106:
Mendelian laws. It is not sufficien
- Page 107 and 108:
example, the problem of the objecti
- Page 109 and 110:
a linear function is not restricted
- Page 111 and 112:
258 - 82 - 176 cases or 68.5% of al
- Page 113 and 114:
The Framingham investigation indeed
- Page 115 and 116:
or, for discrete observations,IT(ω
- Page 117 and 118:
What objections can be made? First,
- Page 119 and 120:
eliability and queuing are known to
- Page 121 and 122:
Kolman E. (1939 Russian), Perversio
- Page 123 and 124:
measurement is provided. Recently,
- Page 125 and 126:
which means that sooner or later th
- Page 127 and 128:
The foundations of the Mises approa
- Page 129 and 130:
A rather subtle arsenal is develope
- Page 131 and 132:
4.3. General remarks on §§ 4.1 an
- Page 133 and 134:
BibliographyAlimov Yu. I. (1976, 19
- Page 135 and 136:
processes are now going on in the s
- Page 137 and 138:
obtaining a deviation from the theo
- Page 139 and 140:
VIOscar SheyninOn the Bernoulli Law