Estimating loss functions of experts

Estimating Loss Functions of Experts

We propose a new and simple methodology to estimate the loss function associated with experts’ forecasts. Under the assumption of conditional normality of the data and the forecast distribution, the asymmetry parameter of the lin-lin and linex loss function can easily be estimated using a linear regression. This regression also provides an estimate for potential systematic bias in the forecasts...

Learning to Rank Using Mixture Of Experts And Matching Loss Functions

Estimating Functions

Estimating functions provide inference methods based on a model for speciied functions of the response variable, such as the mean and variance. The inference methods can use the limiting distribution of the estimating function, or the derived limiting distribution of the estimating function roots, or the derived limiting distribution of the quasi-likelihood function. In fully speciied parametri...

Postoperative visual loss: experts, data, and practice.

The anesthesiologist clearly caused this patient’s vision loss by letting the mean arterial blood pressure drop below 60 mmHg for more than 5 min. This patient is now blind because the anesthesiologist let the patient’s hematocrit level decrease to less than 24% during the case. The person providing the patient’s anesthesia failed to avoid prolonged pressure on the eye, leading to the patient’s...

Spectral Experts for Estimating Mixtures of Linear Regressions

Discriminative latent-variable models are typically learned using EM or gradient-based optimization, which suffer from local optima. In this paper, we develop a new computationally efficient and provably consistent estimator for a mixture of linear regressions, a simple instance of a discriminative latentvariable model. Our approach relies on a lowrank linear regression to recover a symmetric t...