Suchergebnisse

SummaryPrediction of water levels.Predicting of water levels may consist of: a. Prediction of the level on a given moment, b. Prediction of change through hydraulic works. c. Indication of the level to be used as a basis for the design of hydraulic works.As only c is of a statistical nature, most attention is given to it. Examples are given, but when — as in deciding about the height of dikes — the consequences of higher waterlevels than were used as a basis may be disastrous — much reluctance is met to accept this decision as a statistical one. It is shown how nevertheless the statistical approach is the best with regard to the safety of the areas protected by the dikes and how this approach may be used even when circumstances are new (as in the case of the IJssel lake) and no frequency‐distribution based on experience is available.

SummaryPrediction of historical developments.Apart from the thirst for knowledge concerning the past, the urge to see into the future has invariably fulfilled an important function in the historical activity of the mind. Our ancestors in the middle ages, who gave a mystic interpretation to historical events, regarded history purely as a divine plan which would ultimately lead to the end of the world and the Day of Judgment. Their successors of the Renaissance period, after the Middle Ages, looked back to the ancient classical times by cultivating the notion that the historical process constituted a cycle with a continuous rise and fall, which would continue unchanged in the historical future. Modern history then created two notions regarding the future — developed from historical interpretation — which, although widely different, found each other and entered into a close union in the eighteenth and the nineteenth century. These were the notion of History as a Human Plan and the notion of History as Evolution.It is interesting to observe how at the present time, whilst eagerly weighing the chances of our Western future, new prognoses are actually being tried out — with more extensive material than ever before — but also how these prognoses of our time have changed nothing in and have added nothing to the four older systems. In the present era the four themes are powerfully active — completely at random — in our world of thought.

SummaryPrediction of the weather.The daily predictions of the minimum temperature to be expected during the night, which have been published by the Royal Netherlands Meteorological Institute since 21st September, 1952, fulfil the statistical requirements, but on closer inspection it appears that the weak point in these predictions is the tolerance factor. The correlation coefficient of the expected and the observed tolerance during 86 observation days is only + 0.21.By replacing the tolerance factor in the prediction by its average value the Royal Netherland Meteorological Institute at De Bilt attains a correlation coefficient of + 0.884 between the observed and the expected minimum temperature, while the result of the persistency test (tomorrow the same weather as to‐day) is + 0.760.The investigation into variation (fig. 2) indicates the natural differences between the coastal areas and the inland areas; the variation decreases from the coast to the inland areas.The temperature differences between Den Helder and De Bilt (fig. 3, curve H) during the winter (average values for approximately 1000 observation days) increase according as the minimum temperature at De Bilt decreases. This effect is also observed in the summer (1950 and very high temperatures in 1947).The high correlation coefficient attained proves the reliability of the predictions.

SummaryPrediction of production results.The possibility of the prediction of what is going to happen in the future on the basis of what has happened in the past is demonstrated. A simple explanation of the principles of the statistical control chart and of its manupulation in practice is given. By the application of control chart technique the producer is shown to be in a position to discern between two types of causes affecting the quality of his product, viz. mere chance causes which are unavoidable and other causes which might well prove to be avoidable. Statistical methods, of which the control chart is only one example, are recommended as valuable and powerful fools in handling and solving various types of production problems.

SummaryPrediction of harvests.With respect to the prediction of harvests in agriculture a distinction can be made between short‐term estimates and actual long‐term predictions.Harvest estimates are made by the Municipal Advisory Committees for Agricultural Statistics headed by the Burgomasters. These estimates are made either just before (provisional harvest estimate) or immediately after the harvest (definitive harvest estimate). During a considerable part of the year estimates are made with respect to the development and the expected harvest by personnel of the agricultural and horticultural information services, the state of the crops being expressed in a single figure.Long‐term prediction of harvests are frequently based on relationships between the weather factors and the yields. In the Netherlands some investigations have been carried out in this connection by members of the Royal Netherlands Meteorological Institute and of the agricultural and horticultural information services.Predictions with respect to the development of the livestock position, particularly pigs, are altogether of a different nature. Such predictions are made on the strength of monthly data regarding the number of female animals covered, which information is obtained by means of random tests.Predictions are also made in the fisheries industry, particularly with respect to the herring catches landed.

SummaryPrediction in applied psychology.In applied psychology a distinction is needed between implicit and explicit prediction. Predictions, on the basis of test results, of educational or professional success belong to the latter category. The statistical techniques used are generally simple, since the different data are combined into a psychologically understandable structure rather than into a formula, this structure as a whole serving as basis for the prediction.Very often predictions are implied — or even hidden — in more or less conditional or intrinsic statements. This is partly a result of a tradition of caution. But this caution often interferes with the exigency that specific scientific statements be directly testable.As an example of (implicit) prediction of a chronological sequence the results of a research into the effects of war circumstances upon the intelligence level of schoolchildren were quoted. In 1946 a recovery to the normal pre‐war level, within 5 to 10 years, was correctly predicted.

SummaryPrediction in the economic sphere.The technique of prognosis, as an aid in government policy, is of great importance for the general economic prosperity. A description is given of the method by which the Netherlands Central Planning Bureau, as advisor to the government, endeavours to compile an overall picture of the economic future. Further, information is furnished regarding applied aggregation, in which the micro‐economic magnitudes are combined into macro‐economic variables and the micro‐economic relationships into macro‐economic relationships. The optimum aggregation combines a still adequate wieldability with an acceptable error.Part of the variables in the econometric model must be estimated independently. Generally speaking, the magnitudes can be classified into two categories: the autonomous and the non‐autonomous variables.In addition, there is another important classification, viz. the division into: a. objects (e.g. employment, actual wage level), b. instruments (e.g. taxation, government expenditure), c. data (e.g. growth of population, world‐wide economic conditions), d. irrelevant variables (e.g. nominal wage level, price level).Furthermore, Goudr***aan's criticism of the method for determining the economic structure is discussed, because he is of opinion that the approach to econometry has so far been incorrect in principle, both as regards the analysis of the historical time series and the technique of prognosis.

SummaryEstimation of dates of delivery by extrapolation of progress‐data for a project.The problem can be put in mathematical form with the aid of some assumptions with regard to the occurrence of interruptions.The progress of the project is supposed to be interrupted by time‐intervals during which progress is fully stopped. These intervals start at irregularly distributed moments and have a variable size. The points on the time axis corresponding with the moments of stagnation will show a constant average density, if a constant average production capacity or development capacity is available. This means that the results are valid only if no alterations are made which influence the general progress of the project involved.The application of probability theory to problems connected with promises of delivery may, in general, look unacceptable.Many projects, however, have an experimental character and unforeseen technical and organisational troubles are unavoidable.In such cases the theory makes it possible to obtain a maximum amount of information contained in the progress‐data and may help management to make the best decisions.

Online content related to sex and sexuality is widely available. Both the amount as well as the variety of Internet sexuality have been growing since the nineties. However, little is known about adolescents' characteristics that may lead them to seek out more sexual content online. Using the media practice model as founda¬ tion, this study examines the relationship between sexual attitudes and Internet use for 1,581 university students in Flanders. Regression analyses reveals that students with positive attitudes towards masturbation, who find pleasure more important than love and who feel unsecure about performing sexually more frequently screen the Internet for sexual contents. Moreover, the relationship between attitudes towards masturbation and reporting more uncertainty about sex and searching for sexual contents on the Internet was moderated by gender. This study underscores the need for more comprehensive, longitudinal studies of adolescents' consumption of sexual content online and its connections to sexual development in youth.

SamenvattingBezien wij het probleem van economische voorspelling. Wij nemen aan dat dit met een model gebeurt. Over dit model wordt de volgende veronderstelling gemaakt: Het model is een lineair stelsel van stochastische differentie‐vergelijkingen. Om een voorspelling te kunnen doen met behulp van het model, hebben wij cijfers nodig omtrent de uitgangstoestand. Deze uitgangstoestand behoort tot het recente verleden, waaromtrent de statistische informatie nog on‐volledig is. Dit doet het volgende probleem rijzen: Kan men het model ook gebruiken om de uitgangstoestand te "voorspellen"? Dit probleem verschilt van dat van het voorspellen van de toekomst, doordat we een additionele eis moeten stellen. De "voorspelde" vector van uitgangswaarden moet in overeenstemming zijn met de beperkte statistische informatie die wèl beschikbaar is. De auteur pakt dit probleem aan, door het minimaliseren van een gewogen som van de storingstermen in de afzonder‐lijke relaties van het model. Het algoritme, waartoe dit aanleiding geeft, wordt tamelijk gedetailleerd besproken.