Search

Age, time, and vintage are key determinants of house prices, yet they cannot be included together linearly or as dichotomous variables in hedonic regressions as construction time + age of house = sale time. We introduce a method where the estimates of the age, time, and vintage effects on prices are obtained in a flexible manner, without requiring us to specify a pre-determined functional form for any of these variables. Applying our method to Dutch data, we find that the estimated depreciation pattern over the life of houses does not follow the functional forms typically specified for the age of houses in hedonic regressions. (JEL C43, E01, E31, R31), I. INTRODUCTION It is well known that time of sale, product age, and cohort--three temporal variables--cannot all be included linearly or as dichotomous variables in hedonic regressions because of the [...]

The sale price appraisal ratio (SPAR) method has been applied in a number of countries to construct house price indexes. This paper reviews the statistical and index number properties of the SPAR approach. Three types of SPAR indexes are distinguished: a weighted index, which aims at tracking the price change of the stock of owner-occupied houses, and two unweighted indexes, an arithmetic one and a geometric one. We also discuss stratified versions of these indexes and show how stratification can be used to estimate an expenditure-based Laspeyres-type price index. Empirical results for the Netherlands are given and compared with the repeat-sales index published by the Dutch land registry until January 2008. Keywords: House price index. SPAR method, appraisal value, house price DOI 10.3233/JEM-2009-0317

This paper has two aims. First, it shows how the time dummy method to adjust for quality changes fits into the methodology of statistical agencies applying a geometric mean price index formula at the elementary aggregation level. Second, the paper argues that the ordinary or direct time dummy approach cannot cope with systematic price effects of new and disappearing products. Several 'indirect' alternatives are discussed in which the time dummy coefficient serves as a common adjustment factor and in which systematic effects of new or disappearing products are explicitly taken into account. Special attention is paid to the role of the sampling design, in particular proportional to expenditure sampling. Keywords: Consumer price index, hedonic regression, quality adjustment, sampling

This paper compares two model-based multilateral price indexes: the time-product dummy (TPD) index and the time dummy hedonic (TDH) index, both estimated by expenditure-share weighted least squares regression. The TPD model can be viewed as the saturated version of the underlying TDH model, and we argue that the regression residuals are "distorted toward zero" due to overfitting. We decompose the ratio of the two indexes in terms of average regression residuals of the new and disappearing items. The decomposition aims to explain the conditions under which the TPD index suffers from quality-change bias or, more generally, lack-of-matching bias. An example using scanner data on packaged men's T-shirts illustrates our framework. [ABSTRACT FROM AUTHOR]

Abstract Background Cooperatives delivering out of hours care in the Netherlands are hesitant about the use of expert systems during triage. Apart from the extra costs, cooperatives are not sure that quality of triage is sufficiently enhanced by these systems and believe that call duration will be prolonged drastically. No figures about the influence of the use of an expert system during triage on call duration and triage decisions in out of hours care in the Netherlands are available. Methods Electronically registered data concerning call duration and triage decisions were collected in two cooperatives. One in Tilburg, a cooperative in a Southern city of the Netherlands using an expert system, and one in Groningen, a cooperative in a Northern city not using an expert system. Some other relevant information about the care process was collected additionally. Data about call duration was compared using an independent sample t-test. Data about call decisions was compared using Chi Square. Results The mean call time in the cooperative using the TAS expert system is 4.6 minutes, in the cooperative not using the expert system 3.9 minutes. A significant difference of 0.7 minutes (0.4 – 1.0, 95% CI) minutes. In the cooperative with an expert system a larger percentage of patients is handled by the assistant, patients are less often referred to a telephone consultation with the GP and are less likely to be offered a visit by the GP. A quick interpretation of the impact of the difference in triage decisions, show that these may be large enough to support the hypothesis that longer call duration is compensated for by less contacts with the GP (by telephone or face-to-face). There is no proof, however, that these differences are caused by the use of the triage system. The larger amount of calls handled by the assistant may be partly caused by the fact that the assistants in the cooperative with an expert system more often consult the GP during triage. And it is not likely that the larger amount of home visits in Groningen can be attributed to the absence of an expert system. The expert system only offers advice whether a GP should be seen, not in which way (by consultation in the office or by home visit). Conclusion The differences in call times between a cooperative using an expert system and a cooperative not using an expert system are small; 0.4 – 1.0 min. Differences in triage decisions were found, but it is not proven that these can be contributed to the use of an expert system.