Web data often manifest high levels of uncertainty. We focus on categorical Web data and we represent these uncertainty levels as firstor second-order uncertainty. By means of concrete examples, we show how to quantify and handle these uncertainties using the BetaBinomial and the Dirichlet-Multinomial models, as well as how take into account possibly unseen categories in our samples by using th...

The inductive size bias coupling technique and Stein’s method yield a Berry-Esseen theorem for the number of urns having occupancy d ≥ 2 when n balls are uniformly distributed over m urns. In particular, there exists a constant C depending only on d such that sup z∈R |P (Wn,m ≤ z)− P (Z ≤ z)| ≤ C σn,m 1 + ( n m )3 for all n ≥ d and m ≥ 2, where Wn,m and σ 2 n,m are the standardized count and va...

While much attention has been given to reputation mechanisms to deter fraudulent behavior in online auctions, we know little about the nature of disputes or the types of problems that users face. This is an exploratory analysis of 129 disputes for six different types of products that were sold on eBay. In the descriptive and multinomial analysis, we find eight different types of disputes includ...

The paper summarizes existing hard evidence concerning the changing value of human capital in Slovenia’s transition to a market system. It investigates changes in patterns of job mobility (via estimating multinomial logit and hazard models) and changes in the structure of wages (via earnings functions) associated with education, experience, gender, and ethnicity. Data are drawn from unusually r...

Probability intervals are imprecise probability assignments over elementary events. They constitute a very convenient tool to model uncertain information : two common cases are confidence intervals on parameters of multinomial distributions built from sample data and expert opinions provided in terms of such intervals. In this paper, we study how probability intervals can be transformed into ot...

Conditional inference plays a central role in logical and Bayesian reasoning, and is used in a wide range of applications. It basically consists of expressing conditional relationship between parent and child propositions, and then to combine those conditionals with evidence about the parent propositions in order to infer conclusions about the child propositions. While conditional reasoning is ...

SUMMARY. Rao (1945) proposed a method, based on Fisher's information matrix, for measuring distance between distributions of a parametric family satisfying certain regularity conditions. In this paper, Rao's (1945) method is applied to obtain the distance between two negative multinomial distributions. Some other properties are discussed too.

We present a maximum entropy classifier for cross-lingual pronoun prediction. The features are based on local sourceand target-side contexts and antecedent information obtained by a co-reference resolution system. With only a small set of feature types our best performing system achieves an accuracy of 72.31%. According to the shared task’s official macroaveraged F1-score at 57.07%, we are amon...

This paper investigates the finite-sample behaviour of sovereign credit migration estimators and analyzes the properties of the rating process. Through bootstrap simulations, we compare a discrete multinomial estimator and two continuous hazard rate methods which differ in that one neglects time heterogeneity in the rating process whereas the other accounts for it. The study is based on Moody’s...

This note further explores the Dirichlet-tree distribution developed by Dennis (1991). An inclusion relationship is given for different tree structures and an independence relationship is given for probabilities in the tree. The posterior, evidence, and predictive density are derived for multinomial observations.