Fitting Generalized Linear STAP models
1 2 3 4 5 6 7 | stap_glm.fit(y, z, dists_crs, u_s, times_crs, u_t, weight_functions,
stap_data, max_distance = max(dists_crs), max_time = max(times_crs),
weights = rep(1, NROW(y)), offset = rep(0, NROW(y)),
family = stats::gaussian(), ..., prior = normal(),
prior_intercept = normal(), prior_stap = normal(), group = list(),
prior_theta = list(theta_one = normal()), prior_aux = cauchy(location
= 0L, scale = 5L), adapt_delta = NULL)
|
y |
n length vector or n x 2 matrix of outcomes |
z |
n x p design matrix of subject specific covariates |
dists_crs |
(q_s+q_st) x M matrix of distances between outcome observations and built environment features with a hypothesized spatial scale |
u_s |
n x (q *2) matrix of compressed row storage array indices for dists_crs |
times_crs |
(q_t+q_st) x M matrix of times where the outcome observations were exposed to the built environment features with a hypothesized temporal scale |
u_t |
n x (q*2) matrix of compressed row storage array indices for times_crs |
weight_functions |
a Q x 2 matrix with integers coding the appropriate weight function for each STAP |
stap_data |
object of class "stap_data" that contains information on all the spatial-temporal predictors in the model |
max_distance |
the upper bound on any and all distances included in the model |
max_time |
the upper bound on any and all times included in the model |
weights |
weights to be added to the likelihood observation for a given subject |
offset |
offset term to be added to the outcome for a given subject |
family |
distributional family - only binomial gaussian or poisson currently allowed |
... |
optional arguments passed to the sampler - e.g. iter,warmup, etc. |
prior, prior_intercept, prior_stap, prior_theta, prior_aux |
see |
group |
list of of group terms from |
adapt_delta |
See the adapt_delta help page for details. |
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.