Duration on the Housing Market: A Nonparametric Approach

Abstract: This study examines the relationship between the time-on-the-market and the price of apartments on the housing market in Moscow. It is based on a data set including over 10 000 property objects during the time period January 2012 - October 2014. Only secondary proper-ty is considered. The problem of right-censoring is present in the sample, which means that it is not possible to use the duration data without preliminary adjustment. A nonparamet¬ric approach is employed to estimate the duration for the observations which are censored. First, an estimation of the hazard rate function is obtained. Second, based on the hazard rate the values of censored durations are restored with the help of simulation techniques. The duration data adjusted for censoring is used to examine if the market duration depends on the housing price and type of the seller. The duration data shows to exhibit a positive duration depen-dence. For the Moscow property market, a positive relationship between price and time-on-the-market is found. On average it takes longer time to sell the property with the help of real estate agents than without it.