Noise trading has been intensively studied in finance, but rarely in real estate. Theories of price dispersion have also been well established in retailing research, but less so in real estate. This paper is the first attempt to study the effect of noise trading on price dispersion in the real estate spot and presale (forward) markets. Quality-controlled price dispersion data series are estimated using a sample of transaction data in the housing presale and spot markets in Hong Kong. Our results show that transaction volume has a negative and significant effect on price dispersion in the spot market, but a positive and significant effect in the presale market. These support our conjecture that there are more noise traders in the presale market due to lower transaction costs. The volume effects also provide support for the use of a volume weighted least squares model when constructing a repeat sales index.