A numerical approach to a quantitative problem generally involves making some approximations, often at various points in an iterative method. The method steps can then be translated into lines of code. Usually, more iteration results in a better approximation. (Thankfully, computers don't mind tedious repetition. ☺)

NumPy is a popular and well-maintained Python module for numerical computation. It covers most of the standard numerical methods and also provides the ndarray data structure for storing, slicing, and manipulating multi-dimensional data. Calling the multiply() routine, for instance, will perform either regular multiplication or element-wise multiplication depending on whether the ndarray arguments are scalars, vectors, matrices, etc. Low-level languages like C/C++ or FORTRAN are better for specialized, heavy-duty computation, but NumPy is solid overall and ideal for prototyping and data analysis.