UNIGE document Working paper
previous document  unige:125063  next document
add to browser collection

Phase transition unbiased estimation in high~dimensional settings

Year 2019
Abstract An important challenge in statistical analysis concerns the control of the finite sample bias of estimators. This problem is magnified in high dimensional settings where the number of variables p diverge with the sample size n. However, it is difficult to establish whether an estimator θˆ of θ0 is unbiased and the asymptotic order of E[θˆ] − θ0 is commonly used instead. We introduce a new property to assess the bias, called phase transition unbiasedness, which is weaker than unbiasedness but stronger than asymptotic results. An estimator satisfying this property is such that E[θˆ] − θ0 2 = 0, for all n greater than a finite sample size n ∗ . We propose a phase transition unbiased estimator by matching an initial estimator computed on the sample and on simulated data. It is computed using an algorithm which is shown to converge exponentially fast. The initial estimator is not required to be consistent and thus may be conveniently chosen for computational efficiency or for other properties. We demonstrate the consistency and the limiting distribution of the estimator in high dimension. Finally, we develop new estimators for logistic regression models, with and without random effects, that enjoy additional properties such as robustness to data contamination and to the problem of separability.
Keywords Finite sample biasIterative bootstrapTwo-step estimatorsIndirect inferenceRobust estimationLogistic regression
Full text
Working paper (350 Kb) - public document Free access
(ISO format)
GUERRIER, Stéphane et al. Phase transition unbiased estimation in high~dimensional settings. 2019 https://archive-ouverte.unige.ch/unige:125063

517 hits



Deposited on : 2019-10-25

Export document
Format :
Citation style :