1: \begin{abstract}

2: Cox-Ingersoll-Ross (CIR) processes are extensively used in state-of-the-art models for the approximative pricing of financial derivatives.

3: In particular, CIR processes are day after day employed to model instantaneous variances (squared volatilities) of foreign exchange rates 

4: and stock prices in Heston-type models and 

5: they are also intensively used to model short-rate interest rates. 

6: The prices of the financial derivatives in the above mentioned models 

7: are very often approximately computed by means of

8: explicit or implicit Euler- or Milstein-type discretization methods based on equidistant evaluations of the driving noise processes.

9: In this article we study the strong convergence speeds of all such discretization methods.

10: More specifically,

11: the main result of this article reveals that

12: each such discretization method achieves at most a strong convergence order of $\delta/2$,

13: where $0<\delta<2$ is the dimension of the squared Bessel process associated to the considered CIR process.

14: In particular, we thereby reveal that

15: discretization methods currently employed in the financial industry 

16: may converge with arbitrarily slow strong convergence rates to 

17: the solution of the considered CIR process.

18: We thereby lay open the need of the development of 

19: other more sophisticated approximation methods which

20: are capable to solve CIR processes in the strong sense 

21: in a reasonable computational time and which thus can not

22: belong to the class of algorithms which use equidistant evaluations of the driving noise processes.

23: \end{abstract}

24: