The numerics branch of mathematics deals with methods for numerical calculations on the computer. A lot of problems cannon be solved accurately (at least without any extravagant expenses). That is why one has to find approximate algorithms for those purposes. Hence, one would also like to know the speed of convergence of these algorithms.
Furthermore, when using the computer for numerical calculations one cannot avoid some other errors – even if a (theoretical) exact algorithm is known: In the computer, not all real numbers can be represented exactly. This is why so-called roundoff errors occur.
Finally, some flawed data (which are generally determined with the help of gauging, and are therefore flawed – at least to a certain extent) and some effects which were ignored when developing the mathematical model of a natural process also have an effect an the obtained result, and one usually wants to know the order of magnitude of the possible error.
The course "Numerical Mathematik I" deals with the error phenomena mentioned above, as well as numerical problems in linear algebra (solving systems of linear equations, calculus of observations, etc.).
The course "Numerical Mathematik II" rather deals with problems from analysis (for instance the numerical solution of differential equations and the fourier transform).
![[DE]](../../../img/lang/german.gif "Deutsche Version"){kind=small}
![[EN]](../../../img/lang/english.gif "English version"){kind=small}