In recent years new tools have been developed in order to investigate multifractal properties of experimental and simulated data. One of them is based on the scaling behaviour of partition function calculated from the maxima of the wavelet transform coefficients. This method seems to be very useful for unfolding in the space-scale halfplane the hierarchical structure of fractal data. Another method proposed in this context is a multifractal generalization of the detrended fluctuation analysis (MF-DFA). Both methods are widely spread and commonly used for estimating the multifractal spectra of signals. We critically examine validity of each method if it is applied to high-frequency financial data and also compare the results obtained for some well-known mathematical multifractals.