Did you try to google first?

NumPy, SciPy and matplotlib are Python libraries that contain basic statistics functions. You're going to need them if you want to do anything serious. Statsmodels could also be useful.

http://docs.scipy.org/doc/numpy/reference/routines.statistics.html

http://docs.scipy.org/doc/scipy/reference/stats.html

http://statsmodels.sourceforge.net/

sure. please Google numpy for detail. Numpy is a scienticfic computation pakage for Python.

Did you try to google first?

NumPy, SciPy and matplotlib are Python libraries that contain basic statistics functions. You're going to need them if you want to do anything serious. Statsmodels could also be useful.

http://docs.scipy.org/doc/numpy/reference/routines.statistics.html

http://docs.scipy.org/doc/scipy/reference/stats.html

http://statsmodels.sourceforge.net/

Apart from the ones already mentioned, try out 'rpy2' which implements quite a bit of R functionality within python. Pretty easy to use, especially if you have experience with R.

http://rpy.sourceforge.net/

thanks people

yeah, i should have googled first :) sorry to bother.

Varun Varma,

More correctly speaking, rpy2 is an interface to R that lets you call R from Python. That is, it is not an implementation of R functionality in Python. You will also need R to be installed in your system to be able to use rpy2.

Dear Kimmo,

Right you are. I should have worded that answer a bit more carefully. Sorry about that.

I think you can use pandas module (http://pandas.pydata.org/) or some wrapper for R. There is also SAGE (www.sagemath.org) - but this software is a huge set of packages and not all would be useful for You.

Use pandas to manipulate your data; don't use Numpy directly.  You can still use numpy functions on pandas objects.

Pandas objects have basic stats builtin.  For example: data.mean() or data.cov() will give you the mean and covariance in one line.

There are more stats routines lurking in scipy.stats: http://docs.scipy.org/doc/scipy/reference/stats.html

More complex statistical models are available in the pystatsmodels package, which is going to play nicely with pandas.

So in summary:

    pandas to store/manipulate/plot data

    scipy.stats and statsmodels packages to do the statiscis.