pyhdfe.Algorithm

class pyhdfe.Algorithm

Algorithm for absorbing fixed effects. Class attributes contain counts of observations and fixed effect dimensions, and if computed, singletons and degrees of freedom used by the fixed effects.

An algorithm is initialized by create() with one or more dimensions of fixed effects specified by ids. Once initialized, Algorithm.residualize() absorbs the fixed effects into a matrix and returns the residuals from a regression of each matrix column on the fixed effects.

observations

Number of observations in the data (i.e., the number of rows in ids).

Type

int

dimensions

Number of fixed effect dimensions (i.e., the number of columns in ids).

Type

int

singletons

Number of singleton groups or observations. This will be None if there was no need to identify singletons (i.e., if drop_singletons and compute_degrees were both False in create()).

Type

int or None

singleton_indices

Indices of any singleton observations. This will be None if there was no need to identify singletons.

Type

array or None

degrees

Exact or approximate number of degrees of freedom used by the fixed effects computed according to degrees_method in create(). This will be None if compute_degrees was False in create().

Type

int or None

Examples

Methods

residualize(matrix[, weights])

Absorb the fixed effects into a matrix and return the residuals from a regression of each column of the matrix on the fixed effects.