ISSN:0039-0402    

DOI:10.1111/j.1467-9574.1950.tb00407.x

出版商:Blackwell Publishing Ltd    

年代:1950

数据来源: WILEY

摘要:

SummaryStatistics in South AfricaA survey of the present status of statistics in South Africa treating successively of (I) classical administrative statistics, (2) modern statistics arid industrial applications, (3) teaching, and (4) terminology.

ISSN:0039-0402    

DOI:10.1111/j.1467-9574.1950.tb00408.x

出版商:Blackwell Publishing Ltd    

年代:1950

数据来源: WILEY

摘要:

SummarySome statistical problems in the textile industryPart A. Production control in a spinning mill.It is pointed out that the function of planning is as important as the functioit of inspection. Sound planning requires reliable knowledge of the actual performances of men and machines. As to the machines, the actual amount of drafting of a spinning mill can not be computed from the speed and diameters of the rollers; instead it is necessary to use a working diameter of the roller found by experimental determination of the lengths of the ingoing and outgoing yarn. The amount of variation in outgoing yarn thickness can be computed from the variation in the ingoing yarn, the amount of draft and the amount of doubling. The necessary formulae are developed and illustrated.Part B. Weavers' efficiencyApart from servicing the machines, the weaver has to spend a certain portion of his time in reparing breaks in the yarn. If too many weavers are employed, they will part of their time have nothing to do; if too few, the machines will stand idle, because breaks can not be repaired soon enough. A formula is developed for finding the number of personnel that will reduce losses to a minimum.

ISSN:0039-0402    

DOI:10.1111/j.1467-9574.1950.tb00409.x

出版商:Blackwell Publishing Ltd    

年代:1950

数据来源: WILEY

摘要:

SummaryIs diphtheria avoidable?Inoculation against diphtheria has in the past been practiced with different intensity in different parts of the country; but a direct comparison of percentages of the population falling ill does not provide any trustworthy information as to the effect of inoculation, because an epidemic does not reach its peak everywhere i the same year, and because marked differences in morbidity are known to exist between dgferent groups of the population quite apart from the effect of inoculation.It was observed, however, that inoculation was mainly applied to persons between 5 and 14 years of age, and this fact has been used as a basis for investigating its efficiency. To this end two indices were employed, viz:I1=number of cases in the group below 5 years/number of cases in the group above 14 years andI2number of cases in the group from 5 to 14 years/number of cases in the groups below 5 and above 14 yearsThese were computed separately for different boroughs; I1over a period of about 20 years from I924 onwards, and I2, over the two epidemic years I942 and 1943. The data mentioned in this article refer to 1943 only.The index I1is largely independent of inoculation, because few persons in both the age groups considered were inoculated. Hence I1enables us to investigate what social factors influence the distribution of diphtheria cases amongst dgferent age groups. It was found that density of population and family size were the two main influencing factors, which between them could explain by far the greatest part of the variation in I1(multiple correlation coefficient of 0.94).The index I2depends on inoculation, because inoculation is mainly practiced on the group occurring in the numerator. Hence by comparing the index I2for boroughs where the degree of inoculation was ‘good’, ‘medium’, and ‘bad’, we obtain information concerning the effect of inoculation. The results obtained with I1made it possible, however, to group the boroughs in such a manner that the disturbing effect of population density and family size could be in a large measure eliminated, so that the effect of inoculation could be estimated with much greater certainty.This effect turned out to be highly significant.It is estimated that by regular inoculation the risk of falling ill with diphtheria can be reduced by about 61%, and the risk of death by 78%, while various arguments suggest that these estimates must be too low, and that the real effect of inoculation is higher still. Whether it will be possible by systematic inoculation of the whole population altogether to eliminate diphtheria cannot positiuely be decided, but can not be excl

ISSN:0039-0402    

DOI:10.1111/j.1467-9574.1950.tb00410.x

出版商:Blackwell Publishing Ltd    

年代:1950

数据来源: WILEY

摘要:

SummaryPart A of this paper is devoted to U discussion of the properties of linear operators, briefly called ‘linors’. To the well known series of operatorsΔ= (1,‐2,1), Δ4= (1,‐4,6‐4,1) etc, a series S2= (1,2,1), S4= (1,4,6,4,1) etc. is made to correspond. The series Δ2kis shown to have a detrimental influence on the trend in a time series, whereas the series S2kwill retain the trend but remove random errors and cyclical variations. By combination of Δ‐ and S‐linors a linor is derived which may reasonably be assumed to retain the trend and the cyclical variations but to remove the random errors.Next a simple type of moving average operator is deduced which is adopted as giving the trend alone. By a systematic application of these linors a time series can be split up into a trend, a cyclical vuriation, and a random rest.In part B this analysis is performed with two different economic time‐series, while in part C some general considerutions concerning the degree of regression, the reduction of errors, and the smoothness and fitting of linear oper

ISSN:0039-0402    

DOI:10.1111/j.1467-9574.1950.tb00411.x

出版商:Blackwell Publishing Ltd    

年代:1950

数据来源: WILEY

摘要:

SummaryThe use of unilateral and bilateral critical regions in the testing of hypotheses.This paper endeavours to explain in simple terms the principles of the Neyman‐Pearson theory.Let H0be the hypothesis to be tested. Then the observations availuble for testing H0are first condensed into a single statistic, x, the distribution of which can be evaluated when H0is true. Out of the possible range of values of this statistic a critical region is selected, and H0is rejected when x falls in this region, and not rejected when x falls outside. This critical region is chosen so that(A). the probability of rejecting H0when true has a prescribed upper limit a (or preferably is equal to a);(B). the probability of rejecting H0is higher when an alternative hypothesis H1, is true than when H0itself is true; and(C). if possible, the probability of rejecting H0is a maximum when any hypothesis h out of a set of alternative hypotheses is true.When the set of alternative hypotheses is specified by a single parameter θ, H0corresponding to θ=θ0, these requirements will, under conditions of a general nature, lead to the use of unilateral or of bilateral tail‐errors according to the range of values of θ taken into consideration. If it may be assumed that either θ=θ0or θ=θθ0, the critical regions must be unilateral, but if θ can be both greater or smaller than θ0, they have to be bilateral.The arguments are illustrated by a

ISSN:0039-0402    

DOI:10.1111/j.1467-9574.1950.tb00412.x

出版商:Blackwell Publishing Ltd    

年代:1950

数据来源: WILEY

ISSN:0039-0402    

DOI:10.1111/j.1467-9574.1950.tb00413.x

出版商:Blackwell Publishing Ltd    

年代:1950

数据来源: WILEY

摘要:

SummaryGrowth of the mass production industries has posed new and complex problems in industrial management. Scientific solution of these problems necessitates statistical analysis of the vast quantities of data generated in these industries as a by‐product. To date, selected but limited headway has been made in using these data to solve the problems of industrial management.Improvements bordering on the spectacular have been achieved in selected instances of industrial applications of statistical analysis. Quality control and market research afford two such instances.These improvements have been mainly the result of commencing to use facts at all (to replace hunch) rather than the result of superior analysis of data long utilized. In consequence, these pioneering adventures in industrial applications of statistics have been made largely without the aid of statisticians.However, the great need for science in management, and the presence of vast quantities of data on which to base a science, suggest that industry will use trained statisticians in increasing numbers, to the extent that industry may become a principal employer of statisticians.The professional statistical societies can do much to aid the greater utilization of statistics in industry by:(a). organizing in each society u nrujor division to deal with the problems of statistics in industry.(b). sponsoring joint meetings with societies of managers, industrial engineers, and others interested in industrial statistics.(c). injecting into the current literature and into the statistical text‐books, cases and problems drawn from industry.(d). research to develop new statistical tools, and to simplify existing tools for use by industrial personnel.(e). recognizing useful work in industrial statistics through hono

ISSN:0039-0402    

DOI:10.1111/j.1467-9574.1950.tb00414.x

出版商:Blackwell Publishing Ltd    

年代:1950

数据来源: WILEY

ISSN:0039-0402    

DOI:10.1111/j.1467-9574.1950.tb00415.x

出版商:Blackwell Publishing Ltd    

年代:1950

数据来源: WILEY

摘要:

Book reviewed in this article:Introduction to the theory of probability and statistics, Niels Arley and K. Rander Buch

ISSN:0039-0402    

DOI:10.1111/j.1467-9574.1950.tb00416.x

出版商:Blackwell Publishing Ltd    

年代:1950

数据来源: WILEY