est_regressors | R Documentation |
This function uses the marginal posterior distributions of
the change point locations (estimated by est_changepoints
)
in combination with the conditional (on the change point locations)
posterior distributions of the regressors (estimated by
multinom_TS
) to estimate the marginal posterior
distribution of the regressors, unconditional on the change point
locations.
est_regressors(rho_dist, data, formula, timename, weights, control = list())
rho_dist |
List of saved data objects from the ptMCMC estimation of
change point locations (unless |
data |
|
formula |
|
timename |
|
weights |
Optional class |
control |
A |
The general approach follows that of Western and Kleykamp
(2004), although we note some important differences. Our regression
models are fit independently for each chunk (segment of time), and
therefore the variance-covariance matrix for the full model
has 0
entries for covariances between regressors in different
chunks of the time series. Further, because the regression model here
is a standard (non-hierarchical) softmax (Ripley 1996, Venables and
Ripley 2002, Bishop 2006), there is no error term in the regression
(as there is in the normal model used by Western and Kleykamp 2004),
and so the posterior distribution used here is a multivariate normal,
as opposed to a multivariate t, as used by Western and Kleykamp (2004).
matrix
of draws (rows) from the marginal posteriors of the
coefficients across the segments (columns).
Bishop, C. M. 2006. Pattern Recognition and Machine Learning. Springer, New York, NY, USA.
Ripley, B. D. 1996. Pattern Recognition and Neural Networks. Cambridge University Press, Cambridge, UK.
Venables, W. N. and B. D. Ripley. 2002. Modern and Applied Statistics with S. Fourth Edition. Springer, New York, NY, USA.
Western, B. and M. Kleykamp. 2004. A Bayesian change point model for historical time series analysis. Political Analysis 12:354-374. link.
data(rodents)
document_term_table <- rodents$document_term_table
document_covariate_table <- rodents$document_covariate_table
LDA_models <- LDA_set(document_term_table, topics = 2)[[1]]
data <- document_covariate_table
data$gamma <- LDA_models@gamma
weights <- document_weights(document_term_table)
formula <- gamma ~ 1
nchangepoints <- 1
control <- TS_control()
data <- data[order(data[,"newmoon"]), ]
rho_dist <- est_changepoints(data, formula, nchangepoints, "newmoon",
weights, control)
eta_dist <- est_regressors(rho_dist, data, formula, "newmoon", weights,
control)
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.