\(\newcommand{\B}[1]{ {\bf #1} }\) \(\newcommand{\R}[1]{ {\rm #1} }\) \(\newcommand{\W}[1]{ \; #1 \; }\)
density_table¶
View page sourceThe Density Table¶
Discussion¶
The density table is used to identify densities.
It has the following columns:
density_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.
density_name¶
This column has type text .
This column is unique; i.e., each name can only appear ones in this table.
All of the density_id values that appear in the
prior and
data tables
must appear in the density table.
Below is a list of the possible density names
with a link to the corresponding meaning:
uniform¶
binomial¶
Binomial . This density is special because it can only be used in the data table. The data table meas_std must be null and the sample_size must not be null.
\(n\) |
sample_size in the data table |
\(k\) |
counts in a binomial sample (need not be an integer) |
\(y=k/n\) |
meas_value in the data table |
\(\mu\) |
average integrand is model mean for y. |
\(\sqrt{\mu / n}\) |
approximation used for meas_value standard deviation |
gaussian¶
cen_gaussian¶
log_gaussian¶
cen_log_gaussian¶
laplace¶
cen_laplace¶
log_laplace¶
cen_log_laplace¶
students¶
log_students¶
Notation¶
Linear¶
We refer to the following densities as linear:
gaussian ,
cen_gaussian ,
laplace ,
cen_laplace ,
students .
Censored¶
We refer to the following densities as censored:
cen_gaussian ,
cen_log_gaussian ,
cen_laplace ,
cen_log_laplace .
Log Scaled¶
We refer to the following densities as log scaled:
log_gaussian ,
cen_log_laplace ,
log_laplace ,
log_students .
Nonsmooth¶
We refer to the following densities as nonsmooth:
laplace ,
cen_laplace ,
log_laplace .
Example¶
The file density_table.py
contains an example density table.