Bayesian Binary and Ordinal Logistic Regression
blrm(
formula,
ppo = NULL,
keepsep = NULL,
data,
subset,
na.action = na.delete,
priorsd = rep(100, p),
priorsdppo = rep(100, pppo),
conc = 1/(0.8 + 0.35 * max(k, 3)),
psigma = 1,
rsdmean = if (psigma == 1) 0 else 1,
rsdsd = 1,
ar1sdmean = 1,
iter = 2000,
chains = 4,
refresh = 0,
progress = if (refresh > 0) "stan-progress.txt" else "",
x = TRUE,
y = TRUE,
loo = n a R formula object that can use rms package enhancements such as the restricted interaction operator
formula specifying the model predictors for which proportional odds is not assumed
a single character string containing a regular expression applied to design matrix column names, specifying which columns are not to be QR-orthonormalized, so that priors for those columns apply to the original parameters. This is useful for treatment and treatment interaction terms. For example keepsep='treat' will keep separate all design matrix columns containing 'treat' in their names. Some characters such as the caret used in polynomial regression terms will need to be escaped by a double backslash.
a data frame
a logical vector or integer subscript vector specifying which subset of data whould be used
default is na.delete to remove missings and report on them
vector of prior standard deviations. If the vector is shorter than the number of model parameters, it will be repeated until the length equals the number of parametertimes.
vector of prior standard deviations for non-proportional odds parameters. As with priorsd the last element is the only one for which the SD corresponds to the original data scale.
the Dirichlet distribution concentration parameter for the prior distribution of cell probabilities at covariate means. The default is the reciprocal of 0.8 + 0.35 max(k, 3) where k is the number of Y categories. The default is chosen to make the posterior mean of the intercepts more closely match the MLE. For optimizing, the concentration parameter is always 1.0 to obtain results very close to the MLE for providing the posterior mode.
defaults to 1 for a half-t distribution with 4 d.f., location parameter rsdmean and scale parameter rsdsd
the assumed mean of the prior distribution of the standard deviation of random effects. When psigma=2 this is the mean of an exponential distribution and defaults to 1. When psigma=1 this is the mean of the half-t distribution and defaults to zero.
applies only to psigma=1 and is the scale parameter for the half t distribution for the SD of random effects, defaulting to 1.
the assumed mean of the prior distribution of the standard deviation of within-subject white noise. The setup is the same as with rsdmean.
number of posterior samples per chain for rstan::sampling to run
number of separate chains to run
see rstan::sampling. The default is 0, indicating that no progress notes are output. If refresh > 0 and progress is not '', progress output will be appended to file progress. The default file name is 'stan-progress.txt'.
see refresh. Defaults to '' if refresh = 0. Note: If running interactively but not under RStudio, rstan will open a browser window for monitoring progress.
set to FALSE to not store the design matrix in the fit. x=TRUE is needed if running blrmStats for example.
set to FALSE to not store the response variable in the fit
set to FALSE to not run loo and store its result as object loo in the returned object. loo defaults to FALSE if the sample size is greater than 1000, as loo requires the per-observation likelihood components, which creates a matrix N times the number of posterior draws.
set to a file name to run rstan::pairs and store the resulting png plot there. Set to TRUE instead to directly plot these diagnostics. The default is not to run pairs.
set to 'optimizing' to run the Stan optimizer and not do posterior sampling, 'both' (the default) to run both the optimizer and posterior sampling, or 'sampling' to run only the posterior sampling and not compute posterior modes. Running optimizing is a way to obtain maximum likelihood estimates and allows one to quickly study the effect of changing the prior distributions. When method='optimizing' is used the result returned is not a standard blrm object but is instead the parameter estimates, -2 log likelihood, and optionally the Hession matrix (if you specify hessian=TRUE in ...). When method='both' is used, rstan::sampling and rstan::optimizing are both run, and parameter estimates (posterior modes) from optimizing are stored in a matrix param in the fit object, which also contains the posterior means and medians, and other results from optimizing are stored in object opt in the blrm fit object. When random effects are present, method is automatically set to 'sampling' as maximum likelihood estimates without marginalizing over the random effects do not make sense.
intial value for optimization. The default is the rstan default 'random'. Frequently specifying init=0 will benefit when the number of distinct Y categories grows or when using ppo hence 0 is the default for that.
initial value for sampling, defaults to inito
set to TRUE to return the Stan data list and not run the model
a file name for a saveRDS-created file containing or to contain the saved fit object. If file is specified and the file does not exist, it will be created right before the fit object is returned, less the large rstan object. If the file already exists, its stored md5 hash string datahash fit object component is retrieved and compared to that of the current rstan inputs. If the data to be sent to rstan, the priors, and all sampling and optimization options and stan code are identical, the previously stored fit object is immediately returned and no new calculatons are done.
set to TRUE to output timing and progress information to /tmp/debug.txt
passed to rstan:optimizing. The seed parameter is a popular example.
an rms fit object of class blrm, rmsb, rms that also contains rstan results under the name rstan. In the rstan results, which are also used to produce diagnostics, the intercepts are shifted because of the centering of columns of the design matrix done by blrm. With method='optimizing' a class-less list is return with these elements: coefficients (MLEs), beta (non-intercept parameters on the QR decomposition scale), deviance (-2 log likelihood), return_code (see rstan::optimizing), and, if you specified hessian=TRUE to blrm, the Hessian matrix. To learn about the scaling of orthogonalized QR design matrix columns, look at the xqrsd object in the returned object. This is the vector of SDs for all the columns of the transformed matrix. Those kept out by the keepsep argument will have their original SDs.
Uses rstan with pre-compiled Stan code whose location is given by the user in options(stancompiled='...') to get posterior draws of parameters from a binary logistic or proportional odds semiparametric ordinal logistic model. The Stan code internally using the qr decompositon on the design matrix so that highly collinear columns of the matrix do not hinder the posterior sampling. The parameters are transformed back to the original scale before returning results to R. Design matrix columns re centered before running Stan, so Stan diagnostic output will have the intercept terms shifted but the results of blrm for intercepts are for the original uncentered data. The only prior distributions for regression betas are normal with mean zero, and the vector of prior standard deviations is given in priorsd. These priors are for the qr-projected design matrix elements, except that the very last element is not changed. So if one has a single non-interactive linear or binary variable for which a skeptical prior is designed, put that variable last in the model.
The partial proportional odds model of Peterson and Harrell (1990) is implemented, and is invoked when the user specifies a second model formula as the ppo argument. This formula has no left-hand-side variable, and has right-side variables that are a subset of those in formula specifying for which predictors the proportional odds assumption is relaxed.
blrm also handles single-level hierarchical random effects models for the case when there are repeated measurements per subject which are reflected as random intercepts, and a different experimental model that allows for AR(1) serial correlation within subject. For both setups, a cluster term in the model signals the existence of subject-specific random effects, and an additional model term aTime(time variable) signals the use of the AR(1) within-subject model. The time variable must be integer valued and there can be arbitrary gaps between measurements. However if the maximum time exceeds 200 or so one can expect much longer computation time. When aTime() is present, the cluster-specific random effects then become the random effect for a subject's time=1 record. When aTime is specified, the covariates must not change over records within subject (no time-dependent covariates), as only the covariate vector for the earliest time for a subject is used.
See https://hbiostat.org/R/rms/blrm.html for multiple examples with results.
print.blrm, blrmStats, stanDx, stanGet, coef.rmsb, vcov.rmsb, print.rmsb, coef.rmsb, stanCompile
# NOT RUN {
options(stancompiled='~/R/stan') # need this always
stanCompile() # do this once per computer to compile centrally
getHdata(Titanic3)
dd <- datadist(titanic3); options(datadist='dd')
f <- blrm(survived ~ (rcs(age, 5) + sex + pclass)^2, data=titanic3)
f # model summary using print.blrm
coef(f) # compute posterior mean parameter values
coef(f, 'median') # compute posterior median values
stanDx(f) # print basic Stan diagnostics
s <- stanGet(f) # extract rstan object from fit
plot(s, pars=f$betas) # Stan posteriors for beta parameters
stanDxplot(s) # Stan diagnostic plots by chain
blrmStats(f) # more details about predictive accuracy measures
ggplot(Predict(...)) # standard rms output
summary(f, ...) # invokes summary.rms
contrast(f, ...) # contrast.rms computes HPD intervals
plot(nomogram(f, ...)) # plot nomogram using posterior mean parameters
# Fit a random effects model to handle multiple observations per
# subject ID
f <- blrm(outcome ~ rcs(age, 5) + sex + cluster(id), data=mydata)
# Fit a random effects model that respects serial correlation within
# subject ID
f <- blrm(outcome ~ rcs(age, 5) + sex + cluster(id) + time(visit))
# }
Run the code above in your browser using DataLab