Title: Complex temporal structure of activity in on-line electronic auctions

Abstract: We analyze empirical data from the internet auction site Aukro.cz. The time series of activity shows truncated fractal structure on scales from about 1 minute to about 1 day. The distribution of waiting times as well as the distribution of number of auctions within fixed interval is a power law, with exponents $1.5$ and $3$, respectively. Possible implications for the modeling of stock-market fluctuations are briefly discussed.