Search

Summary The procedure proposed consists in going through the population to be sampled item by item deciding each time with probability p whether the item at hand shall be incorporated in the sample. The 'distances' between successive items in the sample will then form a random sample from a geometric distribution. A series of these random distances can easily be produced on a computer and can be conveniently used for taking the sample required. In some cases this method may have its advantages over the conventional use of a table of random numbers. [ABSTRACT FROM AUTHOR]

Summary The sums of squares associated with the independent variables in a multiple regression equation depend on the order in which these variables are introduced. Two methods have been proposed in the literature to avoid this inconvenience: 'forward selection' or 'backward elimination'. With forward selection the independent variables are introduced in successive stages. The order is not predetermined but at each stage that variable is taken as the next one which produces the highest reduction in the residual sum of squares of the dependent variable. With backward elimination on the other hand, we start with the complete regression equation and eliminate the independent variables from it in the order in which they produce the smallest increases in the residual sum of squares. This paper describes a simple and convenient computational lay-out which can be used for both procedures. In forward selection we start with the matrix of product sums, and in bacward elimination we work from the inverse matrix. In addition these techniques are applied to a variety of practical examples in order to see what results they lead to and what pitfalls may be encountered. [ABSTRACT FROM AUTHOR]

Summary Inaugural Lecture on applied statistics by Professor Hamaker at the Technical University of Eindhoven. A sketch about the role of applied statistics in the development of experimental science. Mijne Heren Curatoren, Mijne Heren Hoogleraren en Adviseurs, Dames en Heren leden van de wetenschappelijke, de technische en de administratieve staven van deze Hogeschool, Dames en Heren Studenten, en voorts Gij alien, die door Uw aanwezigheid van Uw belangstelling in deze plechtigheid blijk geeft, [ABSTRACT FROM AUTHOR]

Summary Theory and practice of sampling by attributes. This paper brings a survey of present day theory and practice of sampling by attributes. In section 2 the main points in the theory of OC-curves are recapitulated. Section 3 lists the various features which people have attempted to incorporate in sampling tables, and the various factors which influence our choice of a sampling plan. Subsequent sections discuss economic theories, the importance of the distribution of the precentage of defective items over the inspection lots, the relation between sample size and lot size, and some of the existing sampling tables. In a final section the possibilities for further improvements in existing sampling practices is considered. One conclusion is that in many situations a constant sample size regardless of lot size would have very definite advantages. [ABSTRACT FROM AUTHOR]

Summary Recent developments in experimental designs for dealing with quantitative factors. This paper reviews some recent developments in experimental designs for dealing with quantitative factors. Successively discussed are: [ABSTRACT FROM AUTHOR]

Summary Statistics and Experiments After a brief survey of the general importance of the statistical design of experiments, two particular examples are discussed in detail, one dealing with a series of 7 simple experiments on the proper conditions for the production of swaged joints, the other being a 3 × 3 × 3 design on emission in radiovalves. These examples provide illustrative material for some problems of a general nature, to which little attention is paid in textbooks on the subject. Often more factors are mentioned than can be conveniently incorporated in a single experiment. One of the secrets of succesful experimentation lies in making a proper choice of those factors which should go into an experiment. Where experimentation is easy and rapid a series of simple two - or three - factor experiments is better than one complex design. Analysis of variance is not always needed; frequently too the data resulting from an experiment indicate that text-book methods of analysis do not apply and the statistician or experimenter should be trained in recognizing such situations and in dealing with them. The paper ends with some general conclusions concerning the present and future status of statistics as an integral part of experimental science. [ABSTRACT FROM AUTHOR]

Summary Design and analysis of industrial experiments This paper is based on a lecture on the 'Design and Analysis of Industrial Experiments' given by Dr O. L. Davies on the 8th of May 1954 to the Industrial Section of the 'Vereniging voor Statistiek'. Production, formulation, and testing are distinguished as three separate fields of chemical activity where experimental designs can be applied, and various numerical examples of such experiments are discussed in detail. They consist of a 23 factorial design, a 24 half replicate and a 25 quarter replicate fractional factorial design, and a three-way classification. In a final section the recent designs developed by Box for the exploration of response surfaces are briefly considered. [ABSTRACT FROM AUTHOR]

Summary Experimental designs in industry. It is pointed out that in industry owing to the greater speed of modem mass-production we have a greater need for simple experimental designs than in agriculture. The paper makes a special and extensive study of the simplest of all designs: the two-way classification. A distinction is made between qualitative and quantitative factors; factors for which we do not, and for which we do know the levels. Three cases must then be distinguished: (1) two qualitative factors, (2) one qualitative and one quantitative factor, and (3) two quantitative factors. To these, different methods of variance analysis must be applied; in (1) the customary analysis according to rows and columns is adequate, but in (2) we can effectively use polynomial regression analysis in one direction, and in (3) we can do so in both directions. Examples of each of these types of analysis are provided. The opinion is expressed that the analysis of variance serves to pick out from among a number of conceivable models, the simplest model that will adequately describe the observations, and further to provide an estimate of the residual fluctuations. The result of an analysis of variance is of no use for technical purposes; the model chosen expressed in the form of a numerical equation is the information to be used in the factory. [ABSTRACT FROM AUTHOR]

Summary Towards efficient experiments. Under industrial conditions, where we encounter large random fluctuations, large numbers of observations, and a multiplicity of influencing factors, a proper analysis of the data on a statistical basis has become an essential part of our experimental techniques, which ultimately leads to more efficient methods of experimentation. This is illustrated by 7 practical examples. The first 4 examples deal with simple experiments for comparing two batches of raw materials when disturbing factors such as a time trend must be eliminated. The last 3 examples discuss more complex cases, namely a three-factor experiment in the foundry, a latin square experiment in nylon spinning, and a five-factor experiment of a mixed kind on gear-noise testing. An attempt is made to explain in a simple form the relation between the statistical analysis of and the technical conclusions to be drawn from the result of a properly designed experiment. [ABSTRACT FROM AUTHOR]

Summary An account is given of the International Statistical Conferences, held in Indiaduring the month of December, 1951. [ABSTRACT FROM AUTHOR]

Summary Cinema accomodation in the Netherlands and the factors by wich it is determined. The demand for cinema accomodation, B, measured by the number of seats limes the number of performances per week, is correlated with the average income, I, the population density, D, and the denomination of the people, G. This last factor is measured by the proportion of members in the provincial councils coming from those political parties which are acting on orthodox protestant principles. The quantities B, I, D, and G are computed for each of the n provinces in the Netherlands, and subsequently a multiple correlation analysis is carried out with the purpose of explaining log B as a junction oj log /, log D and log G. The resultant regression equation reads log B - log 1.48 + 0.40 log /+ 0.56 log D - 0.41 log G, the multiple correlation coefficient amounting to o.c/65. Since I and D are highly correlated (see fig. 1) the coefficients of log I and log D in the above formula are uncertain. This is particularly pointed out in an editorial postscript where the numerical analysis has been carried out to some detail. It is found that by omitting either I or D the multiple correlation coefficient would be reduced from 0.965 only to 0.95 [ABSTRACT FROM AUTHOR]

Summary The 'International Statistical Institute' and its session at Bern from 5 to 10 September. A general survey Concerning the organisation and activities of the 'International Statistical Institute' and the papers contributed and discussed at its 26th session held at Bern from 5 to 10 September 1949. [ABSTRACT FROM AUTHOR]

Summary Criticisms against the 'Hyperbolic Error Distribution'. In a previous article in this journal (Vol. 1, p. 55, 1948) the hyperbolic error distribution was derived by theoretical argument, starting from simple hypotheses. Against this derivation criticisms have been raised from various sides. These are answered in the present paper. Though it is on the one hand acknowledged that an error has slipped into the previous arguments, it is on the other hand maintained that the hyperbolic distribution can be held upright on a slightly altered basis, namely on a set of approximate, instead of exact, equations. In a postscript the editors point out that serious objections can be raised against these altered views. [ABSTRACT FROM AUTHOR]

Summary Physicists and chemists hardly ever use the standard deviation for expressing the errors of their observations. This is attributed to the facts that the error theory as commonly treated in textbooks on physics is based too exclusively on the normal frequency distribution, and leaves some of the most important practical problems unanswered. The main advantage of mean and standard deviation lies in the very simple mathematical relations satisfied by these quantities regardless of the shape of the frequency distribution. These relations are discussed in the first part of this paper, while in the second part they are employed for a thorough treatment of the degree of rounding off permissible in practice. The following rules of thumb are derived: The maximum permissible rounding interval should be [ABSTRACT FROM AUTHOR]

Summary The construction of large-size random samples by means of random sampling numbers in the usual way requires much tedious labour. This paper describes a short-cut technique by which artificial samples, of say 500 or 1000, can be obtained in about one tenth of the time. Use is made of previous records of the frequencies of two-digit random sampling numbers; these are called the random sampling frequencies'. [ABSTRACT FROM AUTHOR]

Summary (Random sampling numbers). This paper describes a simple dice-throwing technique for producing random sampling numbers. The technique has been tested in various ways on series of up to 40.000 throws without showing evidence of bias. [ABSTRACT FROM AUTHOR]

Summary (A systematic comparison of the statistical properties of present-day sampling schemes). A survey of various sampling schemes, carried out with the aid of the, random walk diagram, 'leads to the conception that it must be possible to effect the same degree of inspection by the application of different schemes. The degree of inspection of a scheme is contained in its, operating characteristic', which is specified by two constants viz: its centre q0, for which P = 1/2, and its slope s in this point, defined by s= -( dP/dq) q= q0. It is shown that, if the operating characteristics of two different sampling schemes possess the same values of q0 and s, the two characteristics are almost completely coincident. Thus two sampling schemes having the same q0 and s will give identical inspection performances, and are consequently defined as, equivalent. 'And by comparing the average sample sizes of equivalent schemes we may compute their relative efficiency'. For each set of values q0 and s it is possible to construct a single sampling scheme possessing these characteristics. Hence to any scheme whatever there will correspond an, equivalent single sampling scheme, and, by adopting the single sampling scheme as a standard of reference, we may arrive at a general definition of efficiency. By the introduction of q0 and s the formulae relating to the sequential sampling scheme are greatly simplified, and for each set of values q0 and s there is one corresponding sequential scheme. By discussing a particular example of the double sampling scheme it is shown that the methods developed in this paper can successfully be applied to a systematic treatment of the variety of schemes now in existence. [ABSTRACT FROM AUTHOR]

Summary In the Netherlands all gambling is forbidden except in a few cases specially mentioned in the law. Time and again specialists in probability theory or statistics are called upon to testify as experts in lawsuits against law evaders. This paper discusses the part played by the experts in such legal procedures. It is pointed out that according to Dutch jurisprudence there does not exist an exclusive distinction between games of chance and games of skill. A simple declaration by an expert that a given game is largely based on chance suffices to see the game sentenced as against the law. This and various other related questions are discussed in detail and illustrated by examples taken from personal experience in the courts. [ABSTRACT FROM AUTHOR]