1: \begin{abstract} \noindent

2: Many iterative methods for solving optimization or feasibility problems

3: have been invented, and often convergence of the iterates to some solution is proven.

4: Under favourable conditions, one might have additional bounds on the

5: distance of the iterate to the solution leading thus to 

6: \emph{worst case estimates}, i.e., how fast the algorithm must converge.

7: 

8: Exact convergence estimates are typically hard to come by.

9: In this paper, we consider the complementary problem of finding 

10: \emph{best case estimates}, i.e., how slow the algorithm has to converge,

11: and we also study 

12: \emph{exact asymptotic rates of convergence}. Our investigation focuses on convex feasibility

13: in the Euclidean plane, where one set is the real axis while the other

14: is the epigraph of a convex function. This case study allows us to obtain

15: various convergence rate results. We focus on the popular method of

16: alternating projections and the Douglas--Rachford algorithm. These methods

17: are connected to the proximal point algorithm which is also

18: discussed. Our findings suggest

19: that the Douglas--Rachford algorithm outperforms the method of alternating

20: projections in the absence of constraint qualifications. Various examples

21: illustrate the theory. 

22: \end{abstract}

23: