Estimating the input of a L\'evy-driven queue by Poisson sampling of the workload process

TL;DR

This paper develops a semi-parametric method to estimate the input process of a Le9vy-driven queue using Poisson-sampled workload data, providing consistency, asymptotic normality, and explicit variance formulas.

Contribution

It introduces a novel method-of-moments estimator for the Le9vy process's characteristic exponent based on workload samples, including a partial MLE for subordinator inputs and a bias-controlled estimator for spectrally-positive inputs.

Findings

Estimator is consistent and asymptotically normal.

Explicit formulas for asymptotic variance are derived.

Bias can be minimized by threshold selection.

Abstract

This paper aims at semi-parametrically estimating the input process to a L\'evy-driven queue by sampling the workload process at Poisson times. We construct a method-of-moments based estimator for the L\'evy process' characteristic exponent. This method exploits the known distribution of the workload sampled at an exponential time, thus taking into account the dependence between subsequent samples. Verifiable conditions for consistency and asymptotic normality are provided, along with explicit expressions for the asymptotic variance. The method requires an intermediate estimation step of estimating a constant (related to both the input distribution and the sampling rate); this constant also features in the asymptotic analysis. For subordinator L\'evy input, a partial MLE is constructed for the intermediate step and we show that it satisfies the consistency and asymptotic normality conditions. For general spectrally-positive L\'evy input a biased estimator is proposed that only uses workload observations above some threshold; the bias can be made arbitrarily small by appropriately choosing the threshold.