This study is a comparative analysis of inflation hedging properties of stocks, gold and real estates for the US. It is hypothesised that the assets have varying market characteristics and thus should respond differently to high inflation. The Fisher's hypothesis for asset-inflation hedging is constructed both within the bivariate and multivariate modelling frameworks. Thereafter, some salient features typical of predictive models such as asymmetry, time-variation and structural breaks are incorporated in the estimation process for completeness. The results show that inflation hedging tendencies of assets are heterogeneous across the considered assets. The real estates and stocks prove to be good hedges against inflation, while gold investment defies Fisher's hypothesis. Also, the results are sensitive to the decomposition of data for pre- and post-GFC periods, indicating that asset-inflation hedging relationship for the US is time-varying. The results are robust to alternative data frequencies.