Mixpanel Statistics

A collection of Python scripts that pull API data from Mixpanel and perform statistics on the data.

Currently supported: - Correlation Analysis - Regression Analysis

Setup

Retrieve your API key and secret from Mixpanel API Information (http://mixpanel.com/user/account/#info). Set the variables in your shell:

export MIXPANEL_API_KEY=klsj234kljSLDKFJl243jlksdjf
export MIXPANEL_API_SECRET=lkJSdlkj234lkjsdlfksjdflksjdf

Correlation Analysis

Correlation determines the relationship between two variables. To determine the correlation of different Mixpanel events, do:

./correlation.py [event1] [event2] [event3]...
./correlation.py success_view checkout_view checkout_error


Output
==============================================================================
Correlation coefficients
checkout_view	x	checkout_error:	0.600231
checkout_view	x	success_view:	0.806892
checkout_error	x	success_view:	0.469129

Regression Analysis

Regression analysis studies the relationship between a dependent variable and other independent variables. To perform regression analysis on your Mixpanel events, do:

./regression.py [dependent_var] [independent_var1] [independent_var2]
./regression.py success_view checkout_view checkout_error

Output
==============================================================================
Dependent Variable: success_view
Method: Least Squares
Date:  Sun, 10 Jan 2010
Time:  12:23:16
# obs:                  60
# variables:         3
==============================================================================
variable     coefficient     std. Error      t-statistic     prob.
==============================================================================
const           1.151644      0.767227      1.501047      0.138863
checkout_view           0.049232      0.005862      8.398709      0.000000
checkout_error          -0.032332      0.133101     -0.242911      0.808946
==============================================================================
Models stats                         Residual stats
==============================================================================
R-squared             0.651435         Durbin-Watson stat   2.219304
Adjusted R-squared    0.639205         Omnibus stat         6.593467
F-statistic           53.263875         Prob(Omnibus stat)   0.037004
Prob (F-statistic)    0.000000         JB stat              5.678402
Log likelihood       -160.875422         Prob(JB)             0.058472
AIC criterion         5.462514         Skew                 0.649233
BIC criterion         5.567231         Kurtosis             3.765081
==============================================================================
Regression equation for response variable 'success_view'

success_view = 1.15164371922 + 0.04923(checkout_view) + -0.03233(checkout_error)

Contact

Web: http://bradjasper.com
Twitter: @bradjasper
Email: [email protected]