The Markov and the Mixed‐poisson Models of Library Circulation Compared

Abstract

Two dynamic models of library circulation, the Markov model originally proposed by Morse and the mixed Poisson model proposed by Burrell and Cane, are applied to a large eleven‐year university circulation data set. Goodness of fit tests indicate that neither model fits the data. In both cases, the set of non‐circulating items is larger than that predicted by the model.

A New Frontier of Informetric and Webometric Research: Mining Web Usage Data

“…For example, numerous studies ana- (Burrell & Fenton, 1994;Decroos et al, 1997;Tague & Ajiferuke, 1987) and library shelving data (Duy & Vaughan, 2006). From this perspective, studying Web usage data should be a natural extension of informetric research.…”

Yes, the GIGP really does work—and is workable!

“…However, statistical estimation is surely to be preferred to informal assessment. [If Sichel's reported fit (p = 0.93) (i) University of Saskatchewan Library (Tague & Ajiferuke, 1987) These data have previously been subjected to various forms of analysis by, among others, Ajiferuke and Tague (1990) and Burrell (1990a and b). We include them here as an example with cy -0, y > 0, which corresponds to a gamma mixture and a negative binomial distribution.…”

“…We include them here as an example with cy -0, y > 0, which corresponds to a gamma mixture and a negative binomial distribution. However, the resulting fit is unacceptable according to Library book circulations, University of Saskatchewan (Tague & Ajiferuke, 1987 the usual x2 criterion (p = 0.0003). While it is a fact of statistical life that almost any theoretical model will be found wanting given a large enough sample, more subtle analyses have suggested that Poisson mixtures are not appropriate for modeling book circulation, with Gelman and Sichel (1987) and Burrell (1990~) arguing for mixtures of binomial and negative binomial, respectively.…”

Using data from computerized circulation and cataloging systems for management decision making in public libraries

“…Model-based circulation forecasts have been proposed, tested, and debated for over 20 years (Bagust, 1983;Burrell, 1980Burrell, , 1982Burrell, , 1985Burrell, , 1986Burrell, , 1987Burrell, , 1988Burrell, , 1990Burrell & Cane, 1982;Chen, 1976;Gelman & Sichel, 1987;Morse, 1968;Tague & Ajiferuke, 1987). These methods fit past circulation data to well-understood mathematical distributions (e.g., exponential, binomial) and then use the mathematical models to forecast circulation.…”