摘要:

Statistics play an important role in the affairs of state. The development over the last 70 years of national economic and product accounts, the creation in 1946 of a Council of Economic Advisers, and advancements in the collection and analysis of economic data have strengthened the ability of policy makers to understand the forces that affect the U.S. economy, anticipate economic changes, and take steps to minimize the adverse effects of market fluctuations. Although it is often correctly said that the business cycle has not been repealed, it is also true that in the last half-century measurement tools have helped produce a remarkable record of virtually uninterrupted economic growth in the United States. Yet problems remain, particularly because data in the social sciences are inherently more prone to error than data in the natural sciences. Perhaps most important, errors in measuring consumer prices can have enormous consequences for the economy and body politic, because the Consumer Price Index (CPI) is used to index Social Security benefits as well as a wide range of federal retirement payments and the tax code. Small differences in measurement have large consequences for federal finances. For example, the Advisory Commission to Study the Consumer Price Index established by the Senate Finance Committee in 1995 estimated that the CPI overstates changes in the cost of living by approximately 1.1 percentage points per year. Correcting this error would “save” the Federal government roughly $1 trillion over 12 years. This measurement error, along with other factors, has contributed to a huge transfer of resources from younger to older age cohorts. However, it is difficult to fully assess the consequences of this resource transfer, because data are not collected or well organized for this purpose. This article suggests that the creation of a Council of Social Advisers, akin to the President's Council of Economic Advisers, could improve the ability of the Federal government to better address the nation's social problems. The article also points out that despite repeated attempts over a period of many years to coordinate the collection, analysis, and use of statistical information by the U.S. government, the Federal statistical infrastructure remains poorly organized. The article also suggests that progress in addressing this continuing problem could be made through enactment of legislation to establish a 15-member commission to study the Federal Statistical System and make recommendations for its improvement. The author, along with other members of Congress, first introduced such legislation in Congress in 1996.

ISSN:0162-1459    

DOI:10.1080/01621459.1999.10474126

出版商:Taylor & Francis Group    

年代:1999

数据来源: Taylor

摘要:

Predicting changes in individual customer behavior is an important element for success in any direct marketing activity. In this article we develop a hierarchical Bayes model of customer interpurchase times based on the generalized gamma distribution. The model allows for both cross-sectional and temporal heterogeneity, with the latter introduced through the component mixture model dependent on lagged covariates. The model is applied to personal investment data to predict when and if a specific customer will likely increase time between purchases. This prediction can be used managerially as a signal for the firm to use some type of intervention to keep that customer.

ISSN:0162-1459    

DOI:10.1080/01621459.1999.10474127

出版商:Taylor & Francis Group    

年代:1999

数据来源: Taylor

摘要:

We explore and illustrate the use of time series decomposition methods for evaluating and comparing latent structure in nonstationary electroencephalographic (EEG) traces obtained from depressed patients during brain seizures induced as part of electroconvulsive therapy (ECT). Analysis of the patterns of change over time in the frequency structure of such EEG data provides insight into the neurophysiological mechanisms of action of this effective but poorly understood antidepressant treatment, and allows clinicians to modify ECT treatments to optimize therapeutic benefits while minimizing associated side effects. Our work has introduced new methods of time-frequency analysis of EEG series that identify the complete pattern of time evolution of frequency structure over the course of a seizure, and usefully assist in these scientific and clinical studies. New methods of decomposition of flexible dynamic models provide time domain decompositions of individual EEG series into collections of latent components in different frequency bands. This allows us to explore ECT seizure characteristics via inferences on the time-varying parameters that characterize these latent components, and to relate differences in such characteristics across seizures to differences in the therapeutic effectiveness and cognitive side effects of those seizures. This article discusses the scientific context and problems, development of nonstationary time series models and new methods of decomposition to explore time-frequency structure, and aspects of model fitting and analysis. We include applied studies on two datasets from recent clinical ECT studies. One is an initial illustrative analysis of a single EEG trace, the second compares the EEG data recorded during two types of ECT treatment that differ in therapeutic effectiveness and cognitive side effects. The uses of these models and time series decomposition methods in extracting and contrasting key features of the seizure underlying the EEG signals are highlighted. Through the use of these models we have quantified, for the first time, decreases in the dominant frequencies of low-frequency EEG components during ECT seizures. We have also identified preliminary evidence that such decreases are enhanced under the more effective ECTs at higher electrical dosages, a finding consistent with prior reports and the hypothesis that more effective forms of ECT are more effective in eliciting neurophysiological inhibitory processes.

ISSN:0162-1459    

DOI:10.1080/01621459.1999.10474128

出版商:Taylor & Francis Group    

年代:1999

数据来源: Taylor

摘要:

In this article we introduce a notion of depth in the regression setting. It provides the “rank” of any line (plane), rather than ranks of observations or residuals. In simple regression we can compute the depth of any line by a fast algorithm. For any bivariate datasetZnof sizenthere exists a line with depth at leastn/3. The largest depth inZncan be used as a measure of linearity versus convexity. In both simple and multiple regression we introduce the deepest regression method, which generalizes the univariate median and is equivariant for monotone transformations of the response. Throughout, the errors may be skewed and heteroscedastic. We also consider depth-based regression quantiles. They estimate the quantiles ofygivenx, as do the Koenker-Bassett regression quantiles, but with the advantage of being robust to leverage outliers. We explore the analogies between depth in regression and in location, where Tukey's halfspace depth is a special case of our general definition. Also, Liu's simplicial depth can be extended to the regression framework.

ISSN:0162-1459    

DOI:10.1080/01621459.1999.10474129

出版商:Taylor & Francis Group    

年代:1999

数据来源: Taylor

ISSN:0162-1459    

DOI:10.1080/01621459.1999.10474130

出版商:Taylor & Francis Group    

年代:1999

数据来源: Taylor

ISSN:0162-1459    

DOI:10.1080/01621459.1999.10474131

出版商:Taylor & Francis Group    

年代:1999

数据来源: Taylor

ISSN:0162-1459    

DOI:10.1080/01621459.1999.10474132

出版商:Taylor & Francis Group    

年代:1999

数据来源: Taylor

ISSN:0162-1459    

DOI:10.1080/01621459.1999.10474133

出版商:Taylor & Francis Group    

年代:1999

数据来源: Taylor

ISSN:0162-1459    

DOI:10.1080/01621459.1999.10474134

出版商:Taylor & Francis Group    

年代:1999

数据来源: Taylor

ISSN:0162-1459    

DOI:10.1080/01621459.1999.10474135

出版商:Taylor & Francis Group    

年代:1999

数据来源: Taylor