. La revue technique de Bell System . ular forme la distribution théorique avec celle observée. Cette méthode est, cependant, inexacte et (jualitative. Il ne nous fournit pas un quantitativemethod de mesurer la proximité de l'ajustement entre la théorie et la pratique, et dans certains cas, il est absolument trompeur. Il est intéressant de voir comment toutes ces choses peuvent être vraiment dites de l'un et du samemethod. Les deux premières caractéristiques, c'est-à-dire, les plus anciennes et les plus simples, sont peut-être facilement accordées. Il reste à souligner plus 70 BEI.L SYSTÈME TECHNIQUE JOURNAL certainement dans lequel la méthode est
1639 x 1525 px | 27,8 x 25,8 cm | 10,9 x 10,2 inches | 150dpi
Informations supplémentaires:
Cette image peut avoir des imperfections car il s’agit d’une image historique ou de reportage.
. The Bell System technical journal . ular form the theoreticaldistribution with the one observed. This method is, however, inaccurate and (jualitative. It does not furnish us with a quantitativemethod of measuring the closeness of fit between theory and practice, and in certain instances it is absolutely misleading. It is of interestto see how all of these things can be truly said of one and the samemethod. The first two characteristics, that is, oldest and simplest, are perhaps readily granted. It remains to be pointed out more 70 BEI.L SYSTEM TECHNICAL JOURNAL definitely wherein the method is sadly deficient as a quantitativemeasure, and therefore often misleading; whereas in certain instancesit may be, nevertheless, the only practical method that can be useil.Graphical Method. The graphical method itself may be subdividedinto two parts. Let us consider first the plot of the observed andtheoretical frequencies. As an example of the unsatisfactory natureof this form of comparison, it is of interest to consider certain data. r. 2 z 1^ /J z to 6i ^s ^a S7 s /9S izo /ea /OS /a^ , »*» •«5 7 79Z , - - -JWJ 6 f*& fZ-» ;^ 7 7 742 ■j^y e S3S zzc .-■;.■ .-.3 *■*/? IC 7/ 66 GO 79 H il : 2 /& /6 . 7 /Z 2 3 nrma ■40« •tor? J>0/^ .2dt09 :4:fS^jo6Z 1 ffteoerrtA o /*»*. ..•.TS. i.tcoc -ri^TTTi (T-im n^tt TSa t.JCC f..-.?7 t.-MlXl.Kt. a, -re33 4.;«. (L.tiaipi^.^i 3 a % 7 % ^~iur*fBze or Successes Fig. S given by Yule in which 12 dice are thrown 4, 096 times, a throw of4, 5, or G points being reckoned a success. If the dice are symmetricalp = q = Y2. and the theoretical distribution if given by 4, 090 (, 2+}2)-the terms of which as given by Yule are presented in the third columnof Fig. 8. It is suggested that the reader, before going further, consider the graphical and tai)ul, ir representation of these (lat.i.The smooth curve is the theoretical distribution 4, 09(3(3 2+-i)--It has been the authors experience to find that in practically everyinsta
