An outline of the library

The narrative in this section goes into greater detail on how to use the components of Apophenia. You are encouraged to read A quick overview first.

This overview begins with the apop_data set, which is the central data structure used throughout the system. Section Databases covers the use of the database interface, because there are a lot of things that a database will do better than a matrix structure like the apop_data struct.

Section Models covers statistical models, in the form of the apop_model structure. This part of the system is built upon the apop_data set to hold parameters, statistics, data sets, and so on.

Histosec covers probability mass functions, which are statistical models built directly around a data set, where the chance of drawing a given observation is proportional to how often that observation appears in the source data. There are many situations where one would want to treat a data set as a probability distribution, such as using apop_kl_divergence to find the information loss from an observed data set to a theoretical model fit to that data.

Section Tests & diagnostics covers traditional hypothesis testing, beginning with common statistics that take an apop_data set or two as input, and continuing on to generalized hypothesis testing for any apop_model.

Because estimation in the apop_model relies heavily on maximum likelihood estimation, Apophenia's optimizer subsystem is extensive. Optimization offers some additional notes on optimization and how it can be used in non-statistical contexts.

Histosec