摘要:

This article reports on tests of the comparative usefulness of consumer attitudes, stock prices, and some non-attitudinal variables for short-term forecasting of the value and number of motor vehicle sales. Both levels and changes of the variables are analyzed for a variety of time spans between 1952 and 1962. The work suggests that stock prices and some non-attitudinal variables, like the length of the work week, share the predictive ability of consumer attitudes, except perhaps for the early part of the 1952–62 period.

ISSN:0162-1459    

DOI:10.1080/01621459.1964.10480745

出版商:Taylor & Francis Group    

年代:1964

数据来源: Taylor

摘要:

In testing one hypothesis against another, observations may be obtained sequentially. The special feature of the situation considered in this paper is that after the decision to stop observation is made a fixed number of additional observations are available for the terminal decision. The test is then based on the ratio of the likelihoods (at two simple hypotheses) of all the observations (whatever stopping rule is used). Numerical comparisons are made for various cases when the hypotheses concern the mean of a normal distribution with the variance known; these cases include a fixed-sample-size procedure as well as cases where several numbers of additional observations are available after the Wald sequential probability ratio test is used as a stopping rule.

ISSN:0162-1459    

DOI:10.1080/01621459.1964.10480746

出版商:Taylor & Francis Group    

年代:1964

数据来源: Taylor

摘要:

The mathematical model developed by Hansen, Hurwitz, and Bershad for the sampling and response errors of surveys is further analysed and also extended to cover the repetition of the original survey. This extension of the mathematical model provides a framework for the joint application of the two devices traditionally used to measure response variance: interpenetrating samples and re-enumeration. The joint application of these two methods facilitates the estimation of more of the parameters of the mathematical model than any one of the methods would permit. Estimators are developed for several of the parameters and their biases are derived. Empirical results are presented on the basis of a study which was undertaken in connection with the 1961 Census of Population (Canada), and which involved the inter-penetration of 67 pairs of enumerators' assignments followed by switching the assignments and re-enumeration. The coefficients of variation of the estimated parameters are also presented.

ISSN:0162-1459    

DOI:10.1080/01621459.1964.10480747

出版商:Taylor & Francis Group    

年代:1964

数据来源: Taylor

摘要:

This paper presents a unified derivation of a number of parameter-free distributions, some of which have appeared earlier in the literature. The starting point is the information provided by a random sample drawn from a certain population and the frequencies observed in a number of mutually exclusive classes. Several types of populations and several ways of defining the classes are considered. The distributions derived concern the unknown class frequencies and the unknown cumulative class frequencies in the population, or in a second independent sample.

ISSN:0162-1459    

DOI:10.1080/01621459.1964.10480748

出版商:Taylor & Francis Group    

年代:1964

数据来源: Taylor

摘要:

The paper studies the effect on variance of the method of sample selection and the choice of estimator in a two-way stratified population. Three methods of sampling, simple random and one and two-way stratified, are studied in conjunction with estimators which use no population information, one and two sets of marginal frequencies and cell frequencies. It is shown that whether the information in the population frequencies is used in the sample draw or in the estimator, the sampling variances are approximately the same. Hence efficiency can be associated with the method of sample selection and the estimator by the amount of population information used by each.

ISSN:0162-1459    

DOI:10.1080/01621459.1964.10480749

出版商:Taylor & Francis Group    

年代:1964

数据来源: Taylor

摘要:

A model is constructed for an economic time series with a changing seasonal pattern. This represents the seasonal component as itself a stationary time series with almost all of the spectral mass very near to the six (angular) frequencies, (πj/6). The problem of “predicting” this seasonal component at the time pointn – μ(μ positive, zero or negative) is then solved on the basis of a reasonable model for this spectrum. The unspecified parameters are basically the signal to noise ratios and measures of the capacity of the seasonal to change through time. The estimation of this seasonal component can be regarded as a component in a servomechanism (the total mechanism being impossibly difficult to prescribe precisely) designed to control economic fluctuations. This component (a filter) can be adjusted by choice of the parameters so as to achieve some compromise between adequate damping and quick response.

ISSN:0162-1459    

DOI:10.1080/01621459.1964.10480750

出版商:Taylor & Francis Group    

年代:1964

数据来源: Taylor

摘要:

The optimality of the method of least squares is reconsidered when multicollinearity is present. An analysis is presented of the relationship between estimability and identifiability. The criterion of best linear minimum bias is developed, and shown to be equivalent to that of best linear conditionally unbiased estimation subject to complementary (non-estimable) linear restrictions. Imposition of erroneous estimable linear restrictions is shown to lower variances of estimators if and only if it biases them. All these results rely heavily on the use of the generalized inverse of a matrix, for which a new proof of existence and uniqueness is presented from the viewpoint of duality in linear spaces. Finally, estimation by minimum mean square error is proposed, and this is shown to reduce to the least squares method when either (a) regression coefficients have infinite prior variances, or (b) least squares estimators have small sampling variances.

ISSN:0162-1459    

DOI:10.1080/01621459.1964.10480751

出版商:Taylor & Francis Group    

年代:1964

数据来源: Taylor

摘要:

Define the linear functional relationship with parameters, α, β, by μ = α + βv. This paper considers the problem of confidence interval estimation of α and β separately, based on a sample of independent pairs, (xi,yi),i= 1, 2, ···,n, withExi= vi,Eyi= μi. The pairs, (xi,yi), are assumed to be independent (i≠i′), to have a common (unknown) covariance matrix Σ, and to follow the bivariate normal distribution. It is known that under these circumstances (providing instrumental variates are available) one can define a variance ratio with 2 and (n− 2) degrees of freedom,F2,n–2say, as a function of α, β, sample statistics and instrumental variates to yield a confidence region on (α, β) based on

ISSN:0162-1459    

DOI:10.1080/01621459.1964.10480752

出版商:Taylor & Francis Group    

年代:1964

数据来源: Taylor

摘要:

Migration and duration of residence data in the continuous population registers of Copenhagen, Denmark, have been analyzed to ascertain the extent to which repeated migration is a factor in accounting for high mobility rates. The findings, based on the period 1950–1961, support the view suggested by several American studies that stability of residence is characteristic of a large majority of the population. Stability is most typical of the young and the older age groups and least common among those 25–44 years of age, both male and female. But even for the middle age range, at least 70 per cent of the population was stable. The findings also emphasize that the pattern of repeated mobility is most characteristic of a limited segment of the population, a group which shows a particularly strong tendency to move several times over relatively short time spans. This pattern is particularly characteristic of those in the 25–44 year age group. These data emphasize, therefore, that a high degree of residential stability for most of the population is not at all contradictory with high rates of mobility The repeated moves of a small segment of the population inflate the overall mobility rates well beyond what they would be if mobility were based on the number of migrants instead of on the number of moves.

ISSN:0162-1459    

DOI:10.1080/01621459.1964.10480753

出版商:Taylor & Francis Group    

年代:1964

数据来源: Taylor

摘要:

The following criteria for determining best substitute interval estimators of a given class for a population parameter are considered in this paper: (1) minimizing the expected length of the confidence interval (maximizing the effectiveness); (2) minimizing the sum of the mean (absolute) deviations of the upper and lower confidence bounds from the true value (maximizing the effectivity); and (3) minimizing the sum of the mean squared deviations of the upper and lower confidence bounds from the true value (maximizing the efficiency). Modified forms of these criteria can also be applied to one-sided confidence intervals. Both theoretical and practical reasons are given for preferring criterion (3) to the others. A second main point of the paper is that exact confidence intervals, when available, are superior (and often far superior) to approximate ones. Both points are illustrated by a study of confidence bounds, based on one quasi-range, for the standard deviation of a normal population.

ISSN:0162-1459    

DOI:10.1080/01621459.1964.10480754

出版商:Taylor & Francis Group    

年代:1964

数据来源: Taylor