Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Loss aversion is a widely-recognized cognitive bias that can affect house price determination. Existing empirical approaches to quantify its effects require hedonic methods to predict what prices would be in the absence of loss aversion. However, conventional methods for making this counterfactual prediction unrealistically preclude time variation in the hedonic values of housing attributes, thereby exacerbating specification error and reducing the accuracy of point estimates. We show how time-varying hedonic regression (TVHR) can overcome this deficiency by allowing hedonic values to change over time. Applying the method to a dataset of residential transactions from Auckland, New Zealand, that spans 2009 to 2018, we find that loss aversion significantly inflates transaction prices. However, by comparison, existing methods yield imprecise estimates of this premium that are implausibly large and sensitive to the sample period selected. Our findings suggest that THVR is an attractive alternative to conventional approaches to quantifying loss aversion.