1: \begin{abstract}

2: %This paper focuses on two variants of the positivity-preserving  schemes, namely the truncated Euler-Maruyama (EM)  method and the truncated Milstein scheme, applied to stochastic differential equations (SDEs) with positive solutions and super-linear coefficients. The optimal strong convergence rates of the two schemes are obtained under some regularity and integrability conditions.  Moreover,

3: %%the schemes are applied to SDEs  which  widely appear  in the mathematical finance, including the 3/2, and the A\"it-Sahalia models.

4: %the proposed schemes are applied to directly approximate SDEs with super-linear coefficients (3/2 and  A\"it-Sahalia models) and to indirectly approximate SDE with sub-linear coefficients (CIR model) based on the  Lamperti transformation.

5: %The applications demonstrate flexibility of our methods.

6: %Finally, numerical experiments

7: %are provided to verify the effectiveness of the theoretical results.

8: 

9: In this paper, we propose two variants of the positivity-preserving  schemes, namely the truncated Euler-Maruyama (EM)  method and the truncated Milstein scheme, applied to stochastic differential equations (SDEs) with positive solutions and super-linear coefficients.  Under some regularity and integrability assumptions we derive  the optimal strong convergence rates of the two schemes. Moreover, we demonstrate flexibility of our approaches by applying the truncated methods to   approximate SDEs with super-linear coefficients (3/2 and  A\"it-Sahalia models) directly  and also   with sub-linear coefficients (CIR model) indirectly.

10: Numerical experiments

11: are provided to verify the effectiveness of the theoretical results.

12: 

13: \end{abstract}