Primary Censored Distributions Functions (Vectorized)

Functions for computing primary censored distributions for multiple integer delays. More...

Collaboration diagram for Primary Censored Distributions Functions (Vectorized):

Functions
vector	primarycensored_lcdf_vectorized (data int start, data int n, data int dist_id, array[] real params, data real pwindow, data int primary_id, array[] real primary_params)
vector	primarycensored_sone_lpmf_vectorized (data int max_delay, data real L, data real D, data int dist_id, array[] real params, data real pwindow, data int primary_id, array[] real primary_params)
vector	primarycensored_sone_pmf_vectorized (data int max_delay, data real L, data real D, data int dist_id, array[] real params, data real pwindow, data int primary_id, array[] real primary_params)

Description

Functions for computing primary censored distributions for multiple integer delays.

Function Documentation

primarycensored_lcdf_vectorized()

vector primarycensored_lcdf_vectorized	(	data int	start,
		data int	n,
		data int	dist_id,
		array[]real	params,
		data real	pwindow,
		data int	primary_id,
		array[]real	primary_params )

Compute the primary event censored log CDF at integer delays

Uses primarycensored_analytical_lcdf_vectorized() when check_for_analytical_vectorized() is 1, and otherwise calls primarycensored_lcdf() at each delay. No truncation is applied.

Parameters

start	First delay to compute
n	Last delay to compute, and the length of the result
dist_id	Distribution identifier
params	Array of distribution parameters
pwindow	Primary event window
primary_id	Primary distribution identifier
primary_params	Primary distribution parameters

Returns

Vector whose element d is the log CDF at d, for d in start:n. Elements before start are not computed.

Definition at line 424 of file primarycensored.stan.

Here is the call graph for this function:

Here is the caller graph for this function:

primarycensored_sone_lpmf_vectorized()

vector primarycensored_sone_lpmf_vectorized	(	data int	max_delay,
		data real	L,
		data real	D,
		data int	dist_id,
		array[]real	params,
		data real	pwindow,
		data int	primary_id,
		array[]real	primary_params )

Compute the primary event censored log PMF for integer delays up to max_delay

Parameters

max_delay	Maximum delay to compute PMF for
L	Minimum delay (lower truncation point)
D	Maximum delay (upper truncation point), must be at least max_delay + 1
dist_id	Distribution identifier
params	Array of distribution parameters
pwindow	Primary event window
primary_id	Primary distribution identifier
primary_params	Primary distribution parameters

Returns

Vector of primary event censored log PMFs for delays [0, 1] to [max_delay, max_delay + 1].

This function differs from primarycensored_lpmf in that it:

1. Computes PMFs for all integer delays from [0, 1] to [max_delay, max_delay + 1] in one call.
2. Assumes integer delays (swindow = 1)
3. Is more computationally efficient for multiple delay calculation as it reduces the number of integration calls.

The log CDFs at the integer delays come from primarycensored_lcdf_vectorized(), which uses the analytical solution where check_for_analytical_vectorized() allows it.

// Example: Weibull delay distribution with uniform primary distribution
int max_delay = 10;
real L = 0.0;
real D = 15.0;
int dist_id = 3; // Weibull
array[2] real params = {2.0, 1.5}; // shape and scale
real pwindow = 7.0;
int primary_id = 1; // Uniform
array[0] real primary_params = {};
 
vector[max_delay] log_pmf =
  primarycensored_sone_lpmf_vectorized(
     max_delay, L, D, dist_id, params, pwindow, primary_id,
     primary_params
  );

Definition at line 492 of file primarycensored.stan.

Here is the call graph for this function:

Here is the caller graph for this function:

primarycensored_sone_pmf_vectorized()

vector primarycensored_sone_pmf_vectorized	(	data int	max_delay,
		data real	L,
		data real	D,
		data int	dist_id,
		array[]real	params,
		data real	pwindow,
		data int	primary_id,
		array[]real	primary_params )

Compute the primary event censored PMF for integer delays up to max_delay

Parameters

max_delay	Maximum delay to compute PMF for
L	Minimum delay (lower truncation point)
D	Maximum delay (upper truncation point), must be at least max_delay + 1
dist_id	Distribution identifier
params	Array of distribution parameters
pwindow	Primary event window
primary_id	Primary distribution identifier
primary_params	Primary distribution parameters

Returns

Vector of primary event censored PMFs for integer delays 1 to max_delay

This function differs from primarycensored_pmf in that it:

1. Computes PMFs for all integer delays from [0, 1] to [max_delay, max_delay + 1] in one call.
2. Assumes integer delays (swindow = 1)
3. Is more computationally efficient for multiple delay calculations

// Example: Weibull delay distribution with uniform primary distribution
int max_delay = 10;
real L = 0.0;
real D = 15.0;
int dist_id = 3; // Weibull
array[2] real params = {2.0, 1.5}; // shape and scale
real pwindow = 7.0;
int primary_id = 1; // Uniform
array[0] real primary_params = {};
vector[max_delay] pmf =
  primarycensored_sone_lpmf_vectorized(
     max_delay, L, D, dist_id, params, pwindow, primary_id, primary_params
  );

Definition at line 620 of file primarycensored.stan.

Here is the call graph for this function: