Estimating negative binomial demand for retail inventory management with unobservable lost sales

Revenue Management in an Assemble-to-Order Production System

“…Number of periods in the planning horizon T {early,middle,no peak} Table 2: Parameters used in the numerical study with three order types tion because it is commonly used in literature (see, e.g., Ehrenberg (1959) or Agrawal and Smith (1996)) and allows for a high coefficient of variation. In the following, we describe the fixed and varied parameters for the scenarios, which are summarized in Table 2.…”

Stochastic Control and Improvement of Statistical Decisions in Revenue Optimization Systems

“…Application of the Bayesian approach to the censored demand case is given in (Ding et al, 2002;Lariviere & Porteus, 1999). Parameter estimation is first considered in (Conrad, 1976) and recent developments are reported in (Agrawal & Smith, 1996;Nahmias, 1994). Liyanage & Shanthikumar (2005) propose the concept of operational statistics and apply it to a single period newsvendor inventory control problem.…”

Bayesian Approach to Modelling of Quasi-Periodic Intermittent Demand

“…Also, the corresponding inter-arrival-time distribution (Fig. 2) possesses a bi-modal shape, which can not be properly described by the exponential function adopted in the classical negative-binomial model (Agrawal and Smith, 1996;Telang et al 2004).…”

“…The first stochastic demand models were based on the Poison or NBD (negative-binomial) distribution, they originated from research on stochastic interpurchase times and proved to be very accurate in fitting of the aggregated data describing frequently purchased goods (Dunn et al, 1983;Wagner and Taudes, 1987;Gupta, 1991;Agrawal andSmith, 1996, Grange 1998). However, their basic assumption on exponential (or gamma-exponential) distribution of the inter-arrival times does not allow to take into account quasi-periodicity in purchases/visits, which becomes a vital issue for personalization of marketing decisions.…”