Does Time-on-Market Measurement Matter?
Posted: 8 Jan 2015
Date Written: January 5, 2015
Abstract
Time-on-market is one of the most commonly analyzed outcomes in the residential literature. However, existing research provides no consensus agreement on the actual definition of "time-on-market" and its calculation. Very little discussion of the numerous reasonable alternative definitions exists. To address this ambiguity, alternative definitions and related calculations of time-on-market are modeled and across-model differences are evaluated. Second, using multiple listing service data from a large United States city, the importance of relisted properties in defining and assessing time-on-market are modeled and across-model differences are evaluated. Second, using multiple listing service data from a large United States city, the importance of relisted properties in defining and assessing time-on-market is investigated. Results indicate that both time-on-market definition and the handling of relisted properties substantially influence model outcomes and the statistical significance of dependent variables. Results also suggest that much of the existing literature incorporation time-on-market warrants reevaluation using a consistent definition of market time and an adjustment for relisted properties.
Keywords: Time-on-market; Weibull modeling; Hazard functions; TOM; DOM
Suggested Citation: Suggested Citation