Empirical evidence shows that house prices are highly volatile and closely correlated with the business cycle, and the fact is at odds with the evidence that rental prices are relatively stable and almost uncorrelated with the business cycle. To explain the fact, we introduce information heterogeneity into a standard dynamic stochastic general equilibrium (DSGE) model with financial frictions. Agents are endowed with heterogeneous shocks, and rationally extract information from market activities. Since agents are confused by changes in average private signals about future fundamentals, the model generates an amplified effect of technology shocks on house prices, which accounts for the disconnect between house prices and the discounted sum of future rents. In addition, the model provides insights for the lead-lag relationship between residential and nonresidential investment over the business cycle. The solution method developed in this paper can be applied in other DSGE models with heterogeneous information.