Statistics Packages

There are a large number of packages to help with statistics. Two of the most commonly used are NumPy and MatPlotLib.

NumPy and SciPy

NumPy is a vector (array) and matrix based analytical package for Python and is widely used for analysis. NumPy is closely associated, and delivered with, SciPy.

Note: The last time I tried to install SciPy with Python 2.7 it kept generating errors. I finally installed Python 3.3 and was able to install the "scipy-stack" from http://www.lfd.uci.edu/~gohlke/pythonlibs/. Welcome to Python packages! We just installed Wing 6.0 on the lab computers so we may be able to work with it for the external libraries.

You can download NumPy from www.scipy.org. SciPy relies on NumPy so you will want to install it first and then install SciPy if needed.

Fortunately, there are excellent tutorials and extensive examples for NumPy so I will just reference two here:

• Examples - Check out the example on linear regression and others
• Tutorial to get started in NumPy - Start here to learn numpy and scipy in depth

MatPlotLib

MatplotLib is a library for science-based charting and can be downloaded from matplotlib.org.

Note: There are excellent tutorials on the scipy web site are a great way to get started with plotting. However, the imports are no longer correct for the latest versions of Matplotlib as "matplotlib.numerix" needs to change to "numpy". Be ready for similar types of errors as you dive into other packages.