Wavelet-based estimation of power densities of size-biased data

We propose a new wavelet-based method for density estimation when the data are size-biased. More specifically, we consider the power of the density of interest, where this power is some value greater than or equal to half. Warped wavelet bases are employed, where warping is attained by some continuous cumulative distribution function. This can be seen as a general framework for which the conventional orthonormal wavelet estimation is the case with the standard uniform c.d.f. We show that both linear and nonlinear wavelet estimators are consistent, with optimal and/or near-optimal rates. Monte Carlo simulations are performed to compare four special set-ups which are easy to interpret in practice. A real dataset application illustrates the method. We observe that warped bases provide more flexible and better estimates for both simulated and real data. Moreover, we can see that estimating the density power (for instance, its square root) further improves results.