Know<strong>in</strong>g <strong>the</strong> left parts <strong>of</strong> relations (3.2) <strong>and</strong> (3.3) approximatelyequal to <strong>the</strong> frequencies provided by <strong>the</strong> crim<strong>in</strong>al statistics it ispossible <strong>in</strong> pr<strong>in</strong>ciple to determ<strong>in</strong>e both P(A) <strong>and</strong> p. Equations (3.2) <strong>and</strong>(3.3) are <strong>of</strong> a high degree <strong>and</strong> <strong>the</strong>ir solution is not easy. Poisson,however, developed a general method <strong>of</strong> <strong>the</strong>ir solution <strong>and</strong> f<strong>in</strong>allysuccessfully solved <strong>the</strong>m. In that, <strong>the</strong> 19 th century, follow<strong>in</strong>g Laplace<strong>and</strong> Poisson, problems on probabilities <strong>of</strong> verdicts entered alltextbooks on probability <strong>the</strong>ory, but <strong>in</strong> <strong>the</strong> next century suchapplications <strong>of</strong> <strong>the</strong> <strong>the</strong>ory were declared absolutely nonsensical. Weought to f<strong>in</strong>d out <strong>the</strong> reason why.Poisson’s ma<strong>in</strong> presumption was <strong>in</strong>dependence <strong>of</strong> <strong>the</strong> jurors’<strong>in</strong>dividual judgements. Fully underst<strong>and</strong><strong>in</strong>g <strong>the</strong> need to check <strong>the</strong>stability <strong>of</strong> frequencies, he (1837) did not say a word about anexperimental check <strong>of</strong> <strong>in</strong>dependence. How was such a procedurepossible? When solv<strong>in</strong>g equations (3.2) <strong>and</strong> (3.3), Poisson found outthat <strong>the</strong> probability <strong>of</strong> a correct judgement <strong>of</strong> an <strong>in</strong>dividual jurorapproximately equalled 2/3 so that a correct unanimous accusation hadprobability (2/3) 12 < 0.01 <strong>and</strong> was almost impossible. However, <strong>in</strong>neighbour<strong>in</strong>g Engl<strong>and</strong>, as Poisson himself noted, <strong>the</strong> law dem<strong>and</strong>ed aunanimous decision <strong>of</strong> all <strong>the</strong> 12 jurors, <strong>and</strong> English courtspronounced much more condemn<strong>in</strong>g sentences, death sentences<strong>in</strong>cluded, than <strong>the</strong> courts <strong>in</strong> France. To rem<strong>in</strong>d, <strong>the</strong> expositionconcerned <strong>the</strong> 19 th century.Poisson considered that circumstance as a cause for national pride,Engl<strong>and</strong> was seen as a much less civilized nation although it shouldhave been seen as an argument for doubt<strong>in</strong>g his own stochastic model.True, it should be said <strong>in</strong> all fairness that anyway he was unable tocheck it given <strong>the</strong> French crim<strong>in</strong>al statistics. Indeed, protect<strong>in</strong>g <strong>the</strong>secret <strong>of</strong> <strong>the</strong> jurors’ vot<strong>in</strong>g, <strong>the</strong> French judicial code did not dem<strong>and</strong> to<strong>in</strong>dicate <strong>the</strong> number <strong>of</strong> condemn<strong>in</strong>g votes <strong>the</strong> only exception hav<strong>in</strong>gbeen <strong>the</strong> case <strong>of</strong> <strong>the</strong> m<strong>in</strong>imal necessary votes.Thus, from <strong>the</strong> modern viewpo<strong>in</strong>t, Poisson’s error, formallyspeak<strong>in</strong>g, consisted <strong>in</strong> recommend<strong>in</strong>g a stochastic model withoutcheck<strong>in</strong>g it. He determ<strong>in</strong>ed two unknown parameters by two observedmagnitudes with no possibility <strong>of</strong> such check<strong>in</strong>g. It is <strong>in</strong>terest<strong>in</strong>g todescribe <strong>the</strong> pert<strong>in</strong>ent op<strong>in</strong>ion <strong>of</strong> Cournot (1843). Poisson’scontemporary, he apparently was not as ma<strong>the</strong>matically powerful asPoisson, much less as Laplace. However, we ought to recognize tha<strong>the</strong> possessed more common sense <strong>of</strong> a natural scientist, than thosefirst-rate scholars.In particular, he clearly understood that <strong>in</strong>dependence <strong>of</strong> <strong>the</strong> jurors’judgement was only a premise that should have been experimentallychecked. He even proposed such a change <strong>of</strong> <strong>the</strong> judicial code which,without violat<strong>in</strong>g <strong>the</strong> secret <strong>of</strong> <strong>the</strong> jurors’ vot<strong>in</strong>g, would have allowedto obta<strong>in</strong> <strong>the</strong> necessary statistical data. As to <strong>the</strong> <strong>in</strong>dependence itself,Cournot believed that, if it did not exist <strong>in</strong> all <strong>the</strong> totality <strong>of</strong> legalproceed<strong>in</strong>gs <strong>in</strong> general, <strong>the</strong>n <strong>in</strong> any case legislation can be separated<strong>in</strong>to groups <strong>of</strong> <strong>in</strong>dependent cases. He even found out that two suchgroups concern<strong>in</strong>g crimes aga<strong>in</strong>st <strong>the</strong> person <strong>and</strong> aga<strong>in</strong>st property willhave very near to each o<strong>the</strong>r values <strong>of</strong> <strong>the</strong> parameters P(A) <strong>and</strong> p asdeterm<strong>in</strong>ed accord<strong>in</strong>g to <strong>the</strong> Poisson method.28
Nowadays we are sure that no <strong>in</strong>dependence <strong>of</strong> <strong>the</strong> judgement <strong>of</strong><strong>in</strong>dividual jurors does exist, so that <strong>the</strong> groups isolated by Cournotwould have most likely consisted <strong>of</strong> one case only. True, thisstatement is not really proven so that accord<strong>in</strong>g to modern scienceCournot’s po<strong>in</strong>t <strong>of</strong> view is formally <strong>in</strong>vulnerable which once aga<strong>in</strong>confirms that he had essentially outstripped his time.For our days, an important conclusion from <strong>the</strong> above is that it is byno means permissible to use all <strong>the</strong> available statistical <strong>in</strong>formation fordeterm<strong>in</strong><strong>in</strong>g <strong>the</strong> parameters <strong>of</strong> a statistical model; it is absolutelynecessary to leave some part <strong>of</strong> it for check<strong>in</strong>g <strong>the</strong> model itself,o<strong>the</strong>rwise, great scientific efforts can result <strong>in</strong> complete rubbish.4. Substantial Theorems <strong>of</strong> <strong>the</strong> Theory <strong>of</strong> <strong>Probability</strong>4.1. The Poisson <strong>the</strong>orem. When compil<strong>in</strong>g his treatise, Poisson(1837) discovered one <strong>of</strong> <strong>the</strong> ma<strong>in</strong> statistical laws. Calculat<strong>in</strong>g <strong>the</strong>probabilities P(µ = m) that m successes will be achieved <strong>in</strong> n Bernoullitrials, he found out an approximate formula for large values <strong>of</strong> n <strong>and</strong>small values <strong>of</strong> p:λP{µ m}em!m−λ= ≈ (4.1)where λ = np; for more details see Gnedenko (1950). The exactexpression for P{µ = m} depends on three parameters, n, m <strong>and</strong> p; <strong>in</strong><strong>the</strong> approximate expression, n <strong>and</strong> p are comb<strong>in</strong>ed <strong>in</strong>to one.At first sight this simplification seems trivial <strong>and</strong> Poisson himselfdid not th<strong>in</strong>k that his formula was really important. Indeed, his treatise<strong>in</strong>cluded a large number <strong>of</strong> more precise <strong>and</strong> almost as suitableformulas. However, <strong>the</strong> comb<strong>in</strong><strong>in</strong>g mentioned allows to compile acomparatively short table for calculat<strong>in</strong>g (4.1) with two entries, m <strong>and</strong>λ, whereas <strong>the</strong> precise expression for P{µ = m} would have dem<strong>and</strong>eda table with three entries which is not done yet <strong>in</strong> a sufficientlyconvenient form.Never<strong>the</strong>less, <strong>the</strong> ma<strong>in</strong> role <strong>of</strong> <strong>the</strong> formula (4.1) consists not <strong>in</strong>convenient calculation. Strictly speak<strong>in</strong>g, we express it as ama<strong>the</strong>matical <strong>the</strong>orem (Gnedenko 1950) concern<strong>in</strong>g Bernoulli trials, i.e. <strong>in</strong>dependent trials with two outcomes <strong>and</strong> a constant probability <strong>of</strong>success. The most important circumstance is that those conditions maybe violated without deny<strong>in</strong>g its conclusion, that is, <strong>the</strong> equality (4.1).For example, we may admit that different trials have differ<strong>in</strong>gprobabilities <strong>of</strong> successp 1 , p 2 , ..., p n , ...(if only all <strong>of</strong> <strong>the</strong>m are low). Then <strong>the</strong> exact expression for P{µ = m}from Chapter 3 as well as good enough approximate expressionsbecome useless (because <strong>the</strong>y are too exact). The comparatively roughexpression (4.1) rema<strong>in</strong>s valid if only p 1 + p 2 + ...+ p n , or, if desired,np be substituted <strong>in</strong>stead <strong>of</strong> λ. It follows that for calculat<strong>in</strong>g λ it is notnecessary to know <strong>the</strong> values <strong>of</strong> p i , suffice it to know one s<strong>in</strong>gleparameter, <strong>the</strong>ir mean, <strong>the</strong> new value <strong>of</strong> <strong>the</strong> probability <strong>of</strong> success.29
- 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 and 26: where x is any real number. If dens
- Page 27: probability can be coupled with an
- 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