covariate_table

View page source

The Covariate Table

Discussion

The covariate table is used to identify covariate columns in the data table. It has the following columns:

covariate_id

This column has type integer and is the primary key for this table. Its initial value is zero, and it increments by one for each row.

covariate_name

This column has type text and has a different value for every row; i.e., the names are unique and can act as substitutes for the primary key. The names are intended to be easier for a human to remember than the ids.

reference

The column has type real and is the reference value for the corresponding covariate; i.e., the covariate value that corresponds to no adjustment.

max_difference

The column has type real and is the maximum absolute difference, from the reference value, for the corresponding covariate. This value must be greater than or equal zero. The max_difference value null is interpreted as plus infinity and hence no exclusion is done for the corresponding covariate.

Data Subset Table

Rows in the data table for which the covariate corresponding to this covariate_id has a maximum absolute difference greater than max_difference , are not included in the data_subset_table and the fit_data_subset_table .

Avgint Subset

Rows in the avgint_table for which the covariate corresponding to this covariate_id has a maximum absolute difference greater than max_difference , are not included in the predict_table ; see the heading Covariates for the predict table .

Example

The file covariate_table.py contain example covariate table.