Statistical Models for Models

Ben Klemens, a Statistician at U.S. Census Bureau Statistical Research Division, proposed a standardized set of statistical models or framework for agent based models and models in general. According to Klemens the statistical models would then be their own mathematical category.

Klemens presented ‘A Useful Algebraic System of Statistical Models’ at the Computational Social Science department of the Krasnow Institute, where he presented on Oct. 24, 2014.

Some of the highlights:

Klemens’ presentation and research proposes a variety of methods by which researchers can move between data and model parameters.

To learn about the approach in more detail, read Klemens’ paper: “A useful algebraic system of statistical models”.

Epilogue

Klemens also presented on an open source “statistical library for working with data sets and statistical models” that he wrote called Apophenia written in C. From my understanding, Apophnia contains the statistical models that Klemens presented.

According to the github page, “It provides functions on the same level as those of the typical stats package (such as OLS, probit, or singular value decomposition) but gives the user more flexibility to be creative in model-building.”