Statistical property of record breaking events in the Korean housing market

We study the record statistics in the Korean housing market. To characterize the statistical properties of records, we analyze the record rate and the expected number of records for the transaction price of apartments and the volatility. From the numerical analysis, we find that the record rate of price in the overheated region is well described by the record rate obtained from the symmetric independent and identically distributed (IID) random sequence when the price change is relatively small. On the other hand, during the period for the subprime mortgage crisis, the record rate of price in the overheated regions is well approximated by that for the correlated sequence. More interestingly, when the price continuously increases the record rate of price shows a strong seasonal effect in the overheated region, while that in the normal region is simply described by the symmetric IID random sequence with constant drift on the average.