1: \begin{abstract}

2: Let $f$ be analytic on $[0,1]$ with $|f^{(k)}(1/2)|\leq A\alpha^kk!$ for some constants $A$ and $\alpha<2$

3: and all $k\ge1$.

4: We show that the median estimate of $\mu=\int_0^1f(x)\rd x$ under random linear scrambling

5: with $n=2^m$ points converges at the rate $O(n^{-c\log(n)})$ for any

6: $c< 3\log(2)/\pi^2\approx 0.21$. We also get a super-polynomial convergence rate for

7: the sample median of $2k-1$ random linearly scrambled estimates, when $k/m$ is bounded away from zero.

8: When $f$ has a $p$'th derivative that satisfies a $\lambda$-H\"older condition

9: then the median-of-means has error $O( n^{-(p+\lambda)+\epsilon})$

10: for any $\epsilon>0$, if $k\to\infty$ as $m\to\infty$.

11: The proof techniques use methods from analytic combinatorics that have not

12: previously been applied to quasi-Monte Carlo methods, most notably an

13: asymptotic expression from Hardy and Ramanujan on the number of

14: partitions of a natural number.

15: \end{abstract}

16: