1. |
The Draft British Standard 6002: “Sampling Inspection by Variables” |
|
Journal of the Statistical Society of London,
Volume 139,
Issue 3,
2018,
Page 299-308
J. C. Gascoigne,
I. D. Hill,
Preview
|
PDF (785KB)
|
|
ISSN:0959-5341
DOI:10.2307/2344838
出版商:Wiley
年代:2018
数据来源: WILEY
|
2. |
General Discussion on the Draft British Standard 6002 |
|
Journal of the Statistical Society of London,
Volume 139,
Issue 3,
2018,
Page 308-317
Preview
|
PDF (937KB)
|
|
ISSN:0959-5341
DOI:10.1111/j.2397-2327.1976.tb03657.x
出版商:Wiley
年代:2018
数据来源: WILEY
|
3. |
The Ordering of Multivariate Data |
|
Journal of the Statistical Society of London,
Volume 139,
Issue 3,
2018,
Page 318-344
V. Barnett,
Preview
|
PDF (2244KB)
|
|
摘要:
SummaryIn spite of the lack of a natural basis for ordering multivariate data, we encounter an extension of univariate order concepts such as medians, extremes and ranges to the higher dimensional situation. Also, much multivariate theory, and method, exploits order properties in the data or model. We examine the role of ordering in these descriptive and methodological aspects of multivariate analysis by means of a four‐fold classification of sub‐ordering principles.
ISSN:0959-5341
DOI:10.2307/2344839
出版商:Wiley
年代:2018
数据来源: WILEY
|
4. |
Discussion of Professor Barnett's Paper |
|
Journal of the Statistical Society of London,
Volume 139,
Issue 3,
2018,
Page 344-355
Preview
|
PDF (1011KB)
|
|
ISSN:0959-5341
DOI:10.1111/j.2397-2327.1976.tb03659.x
出版商:Wiley
年代:2018
数据来源: WILEY
|
5. |
Computer Studies of Baccarat, II:Baccarat‐Banque |
|
Journal of the Statistical Society of London,
Volume 139,
Issue 3,
2018,
Page 356-364
F. Downton,
Carmen Lockwood,
Preview
|
PDF (587KB)
|
|
摘要:
SummaryBaccarat‐banqueis a game of chance in which a banker playschemin‐de‐fersimultaneously against two players. It is permitted by the Gaming Clubs (Banker's Games) Regulations, 1970. It may be treated as either a two‐person or a three‐person zero‐sum game. This paper describes various optimum strategies for the banker and the players on the assumption that the deck of cards with which the game is played is infinite (or, equivalently, that a finite deck is sampled with replacement).
ISSN:0959-5341
DOI:10.2307/2344840
出版商:Wiley
年代:2018
数据来源: WILEY
|
6. |
Distributed Lag Models for the Quarterly Migration Flows of West Germany, 1962‐72 |
|
Journal of the Statistical Society of London,
Volume 139,
Issue 3,
2018,
Page 365-373
E. G. Drettakis,
Preview
|
PDF (625KB)
|
|
摘要:
SummaryUsing distributed lag models we have investigated in this paper the lags between different components of the (aggregate and by nationality) quarterly migration flows of West Germany during the period 1962/1‐1972/2 and the relation between the flows of the economically active migrants and various indicators of the growth and cyclical fluctuations of the German economy during the same period. In both cases autocorrelation has been fully taken into account.The lag between the aggregate inflow and outflow of the economically active migrants was between three and four quarters while for the inflow from Italy about 6 months. The length of the lag depended on whether a sending country was a member of the EEC, the time her workers started flowing in large numbers into West Germany and the sectoral distribution of these workers in West Germany. For the aggregate inflow it was found that economically active women followed economically active men with a mean lag of less than a month. The lag was about 4 months for the inflows from Turkey and Portugal. The distributed lag models were found less appropriate for the relation between the inflows of the economically non‐active and the economically active migrants. Autocorrelation was significant only in the relation between the economically active women and the economically active men for the inflows from all countries, Italy and Greece.Simple log‐linear regressions of the inflows on the seasonal dummies, the trend and the level of unemployment produced reasonable results. When the Index of Industrial Production was used as an explanatory variable of the inflows of the economically active migrants a distributed lag specification was found appropriate. A double distributed lag with the number of Germans in employment and the outflow of migrant workers as explanatory variables gave significant results for many cases of the inflows of the economically active male and female migrant workers. In most cases in which a distributed lag specification was appropriate the mean lag was about 3 months. The response of the migrants to the changes in the economic conditions of West Germany increased with the difference in theper capitaincome between West Germany and the individual sending countries and with the concentration of the labour force in agriculture in these countries. Autocorrelation was significant for the flows of the countries which started sending migrant workers to West Germany after 1967, i.e. for the flows from Turkey, Portugal and Yugoslavia.
ISSN:0959-5341
DOI:10.2307/2344841
出版商:Wiley
年代:2018
数据来源: WILEY
|
7. |
How the University Grants Committee Determines Allocations of Recurrent Grants—A Curious Correlation |
|
Journal of the Statistical Society of London,
Volume 139,
Issue 3,
2018,
Page 374-384
W. R. Cook,
Preview
|
PDF (623KB)
|
|
摘要:
SummaryThis paper presents the result of a statistical study of the relationship between student enrolment at universities and their receipt of recurrent grants from the U.G.C. It shows an astonishingly high correlation between the actual distribution of recurrent revenue between the United Kingdom universities and entitlements which could be derived from the formula by which the Province of Ontario explicitly determines the entitlement of its universities to recurrent grants. We contrast the uniformity of the allocations of resources to the universities with the variations in the allocations within them to academic departments as shown by the average cost per student in various courses. Taking the results of our own study together with the Bradford study (Bottomleyet al. 1972) of potential cost economies in universities, we show that the nature of the U.G.C.'s distribution formula implies an increasing payment per student in respect of non‐teaching functions as universities get larger. This latter implies one of the following alternatives.(1) Diseconomies of scale in non‐teaching activities due to a higher proportion of expenditure on administration or quasi administration (but contrast Pickford, 1974).(2) A higher payment per student in support of research, quite apart from funds specifically awarded for research.(3) An unusually high level of support for certain privileged areas.In conclusion, we argue the importance of the U.G.C.'s admitting that its block grants are in fact aggregates of its estimates of costs of different groups of students. It would then be possible to examine critically the parameters in the equations. If this, in turn, led to a more precise calculation of the costs of the various areas, and hence to a more logical aggregation, it would not matter if the new improved formula was used by the universities to help them with their internal allocations.
ISSN:0959-5341
DOI:10.2307/2344842
出版商:Wiley
年代:2018
数据来源: WILEY
|
8. |
What is Statistics and Who Should Teach It? |
|
Journal of the Statistical Society of London,
Volume 139,
Issue 3,
2018,
Page 385-388
Freda Conway,
Preview
|
PDF (294KB)
|
|
摘要:
SummaryAt present the organization of the teaching of statistics presents many problems. It would be convenient to regard statistics as applied mathematics which should be taught in mathematics departments, but this solution requires a radical change in the interpretation of the term applied mathematics.
ISSN:0959-5341
DOI:10.2307/2344843
出版商:Wiley
年代:2018
数据来源: WILEY
|
9. |
A Revised Procedure for Dual Record Systems in Estimating Vital Events |
|
Journal of the Statistical Society of London,
Volume 139,
Issue 3,
2018,
Page 389-401
C. C. Greenfield,
Preview
|
PDF (929KB)
|
|
摘要:
SummaryDual record systems for estimating vital events have typically been designed with the intention of correcting their results for bias using an assumption of independence. The operational significance of this assumption and that of other correcting techniques is examined. In the case of independence, it is shown that an underlying condition for the assumption to be correct is for at least one of the two recording systems to be of constant efficiency. However, constant efficiency is ultimately only fully attainable if one of the methods is 100 per cent efficient, which should be an objective in its own right and the attainment of which would make present correcting techniques superfluous. The design of dual record systems to date militates against the achievement of such an outcome and it is suggested that the only way to progress in the desired direction, and to obtain more accurate results in the meantime, is to adopt an efficient sub‐sampling procedure rather than to duplicate records entirely. It is suggested that an appropriate estimator with sub‐sampling is a ratio estimate. An example is given.
ISSN:0959-5341
DOI:10.2307/2344844
出版商:Wiley
年代:2018
数据来源: WILEY
|
10. |
Discrete Multivariate Analysis: Theory and Practice |
|
Journal of the Statistical Society of London,
Volume 139,
Issue 3,
2018,
Page 402-403
R. L. Plackett,
Preview
|
PDF (178KB)
|
|
摘要:
1. Discrete Multivariate Analysis: Theory and Practice.By Yvonne M. M. Bishop, Stephen E. Fienberg and Paul W. Holland. Cambridge, Mass. and London, MIT Press, 1975. x, 557 p.1014″. £15·00.
ISSN:0959-5341
DOI:10.2307/2344845
出版商:Wiley
年代:2018
数据来源: WILEY
|