Empirical tests of forecast optimality have traditionally been conducted under the assumption of mean squared error loss or some other known loss function. This paper establishes new testable properties that hold when the forecaster’s loss function is unknown but testable restrictions can be imposed on the data generating process, trading off conditions on the data generating process against co...

PMC13 ESTIMATING COST-OF-ILLNESS USING GENERALIZED LINEAR MODELS: AN ALTERNATIVETOTHE SMEARING APPROACH Exuzides A, Colby C, Spalding JR ICON Clinical Research, San Francisco, CA, USA, Astellas Pharma US, Deerfield, IL, USA OBJECTIVES: Estimation of cost-of-illness typically involves the analysis of skewed medical costs that include large outliers. Log transformations are frequently used to ove...

We consider the problem of bounding from above the log-partition function corresponding to second-order Ising models for binary distributions. We introduce a new bound, the cardinality bound, which can be computed via convex optimization. The corresponding error on the logpartition function is bounded above by twice the distance, in model parameter space, to a class of “standard” Ising models, ...

This paper considers estimation of normal mean ? when the variance is unknown, using the LINEX loss function. The unique Bayes estimate of ? is obtained when the precision parameter has an Inverse Gaussian prior density

We study the large deviations performance of consensus+innovations distributed detection over random networks, where each sensor, at each time k, weight averages its decision variable with its neighbors decision variables (consensus), and accounts for its new observation (innovation). Sensor observations are independent identically distributed (i.i.d.) both in time and space, but have generic (...

The recent line of study on randomness extractors has been a great success, resulting in exciting new techniques, new connections, and breakthroughs to long standing open problems in the following five seemingly different topics: seeded non-malleable extractors, privacy amplification protocols with an active adversary, independent source extractors (and explicit Ramsey graphs), non-malleable in...

Consider an unknown function L(·) : {1, · · · , d} → {1, · · · , r} with range R = {L(i)|i = 1, · · · , d}. Given d, r, , δ > 0 we show how to compute an estimate p̃ such that with probability at least 1 − δ we have ||R|/r − p̃| ≤ p̃. This is an estimate with a fixed relative error, which is stronger than finding an estimate with a fixed absolute error. This calculation can be performed efficientl...

In this paper the classical estimators of the shape parameter for the Burr Type XII distribution, such as, the Maximum Likelihood Estimator (MLE), the Uniformly Minimum Variance Unbiased Estimator (UMVUE), and the Minimum Mean Squared Error (MinMSE) estimator are obtained. Then the problem of finding the minimax estimators of this parameter under the squared log error, precautionary, and weight...