## SEARCH

#### Country

##### ( see all 186)

- United States 8931 (%)
- Germany 4159 (%)
- Italy 1666 (%)
- Canada 1616 (%)
- France 1537 (%)

#### Institution

##### ( see all 12444)

- University of California 549 (%)
- University of Washington 231 (%)
- Indian Statistical Institute 224 (%)
- The Institute of Statistical Mathematics 211 (%)
- Purdue University 200 (%)

#### Author

##### ( see all 28731)

- Quirk, Thomas J. 133 (%)
- Balakrishnan, N. 124 (%)
- Lehmann, E. L. 118 (%)
- Härdle, Wolfgang Karl 106 (%)
- Janssen, Jürgen 95 (%)

#### Publication

##### ( see all 720)

- Annals of the Institute of Statistical Mathematics 2880 (%)
- Journal of Medical Systems 2316 (%)
- Metrika 2095 (%)
- Statistical Papers 1447 (%)
- Statistics and Computing 1029 (%)

#### Subject

##### ( see all 482)

- Statistics [x] 32044 (%)
- Statistics, general 15675 (%)
- Statistics for Business/Economics/Mathematical Finance/Insurance 14147 (%)
- Statistical Theory and Methods 8490 (%)
- Probability Theory and Stochastic Processes 7330 (%)

## CURRENTLY DISPLAYING:

Most articles

Fewest articles

Showing 1 to 10 of 32044 matching Articles
Results per page:

## Model-based clustering of probability density functions

### Advances in Data Analysis and Classification (2013): 1-19 , June 27, 2013

Download PDF | Post to Citeulike

Complex data such as those where each statistical unit under study is described not by a single observation (or vector variable), but by a unit-specific sample of several or even many observations, are becoming more and more popular. Reducing these sample data by summary statistics, like the average or the median, implies that most inherent information (about variability, skewness or multi-modality) gets lost. Full information is preserved only if each unit is described by a whole distribution. This new kind of data, a.k.a. “distribution-valued data”, require the development of adequate statistical methods. This paper presents a method to group a set of probability density functions (pdfs) into homogeneous clusters, provided that the pdfs have to be estimated nonparametrically from the unit-specific data. Since elements belonging to the same cluster are naturally thought of as samples from the same probability model, the idea is to tackle the clustering problem by defining and estimating a proper mixture model on the space of pdfs. The issue of model building is challenging here because of the infinite-dimensionality and the non-Euclidean geometry of the domain space. By adopting a wavelet-based representation for the elements in the space, the task is accomplished by using mixture models for hyper-spherical data. The proposed solution is illustrated through a simulation experiment and on two real data sets.

## Time and Spatial Series

### Large Sample Techniques for Statistics (2010) 0: 283-315 , January 01, 2010

Download PDF | Post to Citeulike

Time series occur naturally in a wide range of practices. For example, the opening price of a certain stock at the New York Stock Exchange, the monthly rainfall total of a certain region, and the CD4+ cell count over time of an individual infected with the HIV virus may all be viewed as time series.

## An asymptotic test for a geometric process against a lattice distribution with monotone hazard

### Journal of the Italian Statistical Society (1997) 6: 213-231 , December 01, 1997

Download PDF | Post to Citeulike

### Summary

In this paper a simple characterization of the geometric distribution, in the class of discrete distributions with monotone hazard ratio, is provided. This result is used to construct a test for the hypothesis that the anival process of a discrete queueing model is a geometric process. The properties of the test, as well as those of its «bootstrapped version », are studied both theoretically and by Monte Carlo simulation.

## Social Choice

### Mathematics and Politics (2008): 1-48 , January 01, 2008

Download PDF | Post to Citeulike

In the present chapter we consider the situation wherein a group of voters is collectively trying to choose among several alternatives. When people speak of the area of “social choice,” it is typically this context that they have in mind.

In the case where there are only two alternatives, the standard democratic process is to let each person vote for his or her preferred alternative, with the social choice (the “winner”) being the alternative receiving the most votes. The situation, however, becomes complicated if there are more than two alternatives. In particular, if we proceed exactly as we did above where we had two alternatives, then we are not taking advantage of some individual comparisons among the several alternatives that could be made.

## Within-Schätzung bei anonymisierten Paneldaten

### AStA Wirtschafts- und Sozialstatistisches Archiv (2008) 2: 277-297 , October 02, 2008

Download PDF | Post to Citeulike

### Zusammenfassung

Dieser Beitrag untersucht die Auswirkungen der beiden datenverändernden Anonymisierungsverfahren ‚variablenspezifische abstandsorientierte Mikroaggregation‘ (vaabMA, im Englischen: Individual Ranking) und ‚multiplikative stochastische Überlagerung‘ auf die „Within“-Schätzung eines linearen Panelmodells mit Individualeffekten. Es wird gezeigt, dass der „Within“-Schätzer auf der Grundlage der mittels vaabMA anonymisierten Daten konsistent bleibt. Bei der multiplikativen stochastischen Überlagerung wird neben der allgemeinen Form der Überlagerung eine spezielle Variante analysiert, bei der die Variablen zuerst mit einem konstantem Grundüberlagerungsfaktor und danach zusätzlich additiv überlagert werden. Es wird weiterhin gezeigt, dass beide Varianten der Überlagerung zu Inkonsistenz der „Within“-Schätzer führen. Anschließend werden korrigierte Schätzer hergeleitet.

## Optimal and robust invariant designs for cubic multiple regression

### Metrika (1995) 42: 29-48 , December 01, 1995

Download PDF | Post to Citeulike

## Accelerated Life Testing

### The Art of Progressive Censoring (2014): 481-505 , January 01, 2014

Download PDF | Post to Citeulike

Methods of accelerated life testing are applied to several kinds of progressively censored data. This includes step-stress testing as well as progressive stress models.

#### Images from this Article show all 7 images

## On approximating the distribution of quadratic forms in uniform and beta order statistics

### METRON (2013): 1-16 , July 16, 2013

Download PDF | Post to Citeulike

This paper provides a moment-based approximation to the distribution of a quadratic forms in uniform random variables and in order statistics from a uniform population. Certain goodness-of-fit statistics can be expressed in terms of the latter. In particular, it is shown that the proposed methodology yields more accurate percentiles than a previously used approximation in connection with a criterion that is expressible as a quadratic form in order statistics from a uniform distribution. The more general case of quadratic forms in beta random variables is also discussed. The density approximants are expressed as the product of a beta distributed base density function and a polynomial adjustment. Several illustrative examples are provided.

#### Images from this Article show all 13 images

## How reliable are active-control trials?

### evaluating clinical research (2007): 83-88 , January 01, 2007

Download PDF | Post to Citeulike

## Exponential Families

### Theoretical Statistics (2010): 25-38 , January 01, 2010

Download PDF | Post to Citeulike

Inferential statistics is the science of learning from data. Data are typically viewed as random variables or vectors, but in contrast to our discussion of probability, distributions for these variables are generally unknown. In applications, it is often reasonable to assume that distributions come from a suitable class of distributions. In this chapter we introduce classes of distributions called *exponential families*. Examples include the binomial, Poisson, normal, exponential, geometric, and other distributions in regular use. From a theoretical perspective, exponential families are quite regular. In addition, moments for these distributions can often be computed easily using the differential identities in Section 2.4.