1: \begin{abstract}

2: In this paper, we study the approximation and estimation of $s$-concave 

3: densities via R\'enyi divergence. 

4: We first show that the approximation of a probability measure $Q$ by an $s$-concave 

5: density exists and is unique via the procedure of minimizing a divergence functional 

6: proposed by \cite{koenker2010quasi} 

7: if and only if $Q$ admits full-dimensional support 

8: and a first moment. We also show continuity of the divergence functional in $Q$:   

9: if $Q_n \to Q$ in the Wasserstein metric, then the projected densities converge in weighted 

10: $L_1$ metrics and uniformly on closed subsets of the continuity set of the limit. 

11: Moreover, directional derivatives of the projected densities also enjoy local uniform convergence. 

12: This contains both on-the-model and off-the-model situations, and entails strong consistency 

13: of the divergence estimator of an $s$-concave density under mild conditions. 

14: One interesting and important feature for the R\'{e}nyi divergence estimator of an $s$-concave 

15: density is that the estimator

16: is intrinsically related with the estimation of log-concave densities via maximum likelihood methods. 

17: In fact, we show that for $d=1$ at least, 

18: the R\'enyi divergence estimators for $s$-concave densities converge to the maximum 

19: likelihood estimator of a log-concave density as $s \nearrow 0$.

20: The R\'enyi divergence estimator shares similar characterizations as the MLE 

21: for log-concave distributions, which allows us to develop pointwise 

22: asymptotic distribution theory assuming that the underlying density is $s$-concave.

23: \end{abstract}

24: