An empirical study of sample-selection bias in indices of commercial real estate

Real estate price indices based solely on samples of sold properties may no t accurately represent the population of properties due to potential sample -selection bias. This study addresses this potential for sample-selection b ias in the construction of commercial price indices within the context of t he Phoenix area office market. The empirical analysis confirms the presence of sample-selection bias in the estimation of the total price equation. Ho wever, within this sample, the price indices generated after correcting for sample-selection bias do not appear significantly different from those tha t do not consider selectivity bias.