The hardware and bandwidth for this mirror is donated by dogado GmbH, the Webhosting and Full Service-Cloud Provider. Check out our Wordpress Tutorial.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]dogado.de.
MCMCdiag fix spacing issue for min neff in summary file
for cmdstanr objectsMCMCdiag change behavior so that if specified
dir does exist, save to home dir (‘~’)MCMCdiag if mkdir name exists, append ’_1’
to filename and createMCMCsummary that prevented
n.eff from being displayed for jagsUI
objectscmdstanr packageMCMCplot fix bug associated with
ref_ovl = TRUE and the specification of multiple objects.
Function also now automatically lightens colors specified (instead of
defaulting to gray) when ref_ovl = TRUEMCMCsummary fix bug associated with
pg0 = TRUE when input was derived from StanMCMCsummary now takes argument pg0. When
pg0 = TRUE the proportion of the posterior that is greater
than 0 is returned as a columnMCMCdiag now takes any argument from
MCMCsummary to modify printed summary outputMCMCdiag add more descriptive message about output from
nimble objectsMCMCtrace fix rdev matrix warningMCMCplot fix bug that prevented plotting with a single
parameterMCMCdiag function added to create .txt file summarizing
model inputs and outputs and save model object as .rds fileexact as an argument to parse
desired parameter. ISB = TRUE no longer uses regex matching
(exact should be used to specify this).nimble
packageMCMCpstr fix bug where the incorrect number of
dimensions would be returned when subsetting a param with >1 dimMCMCplot now takes the argument ci which
can be used to specify the credible interval displayed in the
caterpillar plotMCMCplot now takes the argument HPD which
can be used to specify whether highest posterior density intervals or
equal-tailed intervals should be displayed in the caterpillar plotMCMCplot fix bug where error would result when trying
to plot a single parameterMCMCplot can now plot models that have NA for posterior
estimates. Useful when plotting two models side-by-side is desired and
one model has parameters that the other does not (use MCMCchains to
create matrix of draws, add NA-filled cols with missing parameter names,
and plot)MCMCtrace fix undesired behavior when
post_zm is specified as FALSEMCMCsummary for model objects fit with
jagsUI, the function now returns Rhat and n.eff values that
coincide with values calculated in that package (e.g., split-chain
Rhat), rather than calculating these with the coda package.
These values are fed into MCMCtrace when Rhat
and/or n.eff are specified as TRUE.MCMCsummary fix bug where an error was returned when
func_name was not specifiedMCMCtrace now accepts expression input (for Greek
characters etc.) from R for main_den and
main_tr argsclass (and use
is instead) as matrix objects will also be
array as of R 4.0.0MCMCtrace no longer changes working dir when
wd is specified - just puts pdf thereMCMCpstr fix bug where dimensions were not displayed
properly for Stan modelsMCMCpstr fix warning with objects derived from
rstanarm and brmsMCMCsummary change variable names to reflect what is
displayed by brms package (rather than raw Stan
output)MCMCsummary fix bug thrown when one only one chain run
for Stan modelsMCMCsummary fix Rhat label issue (from r_hat to
Rhat)MCMCsummary fix bug that caused stanfit parameter names
to be numbersMCMCsummary now takes quantiles as an
argument, allowing user to specify which quantiles to returnMCMCsummary now takes HPD as an argument
to calculate highest posterior density intervalsMCMCsummary now outputs as a data.frame rather than a
matrixMCMCchains fix bug that output different parameters
names for stanreg and brms objectsMCMCplot fix bug that caused guidelines to plot over
appropriate boundsMCMCplot now has object2,
col2, and offset arguments. Allows user to
plot output from two separate models, as long as both model outputs have
identical parameter names.MCMCplot now has guide_lines argument, to
plot lines to help reference which posterior corresponds to each
parameter namerstanarm and
brms packagesMCMCplot add support for plotting multiple colors on
the same plotMCMCsummary for n.eff the default is now
TRUE (displays number of effective samples by default)MCMCtrace now has the option plot. When
FALSE no plot is output. Used in conjunction with
PPO_out = TRUE to to calculate PPO without plotting trace
plots.MCMCsummary fix bug where Stan input was not being
sorted by parameter indexMCMCtrace fix spacing issue for Rhat and n.eff text
when Rhat = TRUE and n.eff = TRUEMCMCplot add deprecation warnings for argumentsMCMCsummary for stanfit objects (model
output derived from Stan) Rhat and n_eff are calculated using the
rstan package. Note that rstan calculates Rhat
and n_eff values slightly differently (more conservatively) than the
coda package (commonly used to summarize model output
derived from JAGS).MCMCplot the following argument names have been
changed: orig: labels_sz, new: sz_labels orig:
med_sz, new: sz_med orig:
thick_sz, new: sz_thick orig:
thin_sz, new: sz_thin orig:
axis_text_sz, new: sz_ax_txt orig:
ax_sz, new: sz_ax orig:
tick_text_sz, new: sz_tick_txt orig:
main_text_sz, new: sz_main_txt orig:
tick_pos, new: pos_tickMCMCtrace add ability to specify xlim and ylim for
density plotsMCMCtrace add ability to specify xlab and ylab for
density and trace plotsMCMCtrace add ability to specify title for trace and
density plotMCMCtrace add ability to specify line width and line
type for density and prior lines on density plotsMCMCtrace add ability to specify color for density and
priors lines on density plotsMCMCtrace add ability to specify size and color of text
when priors specified, and the position of ticks for density and trace
plotsMCMCtrace add ability to specify size of tick labels,
axes labels, and thickness of axesMCMCtrace clean up plotting of trace plots when only
two plots are plotted in windowMCMCtrace add ability to include Rhat and number of
effective samples on trace plotsMCMCchains fix error associated with coda::mcmc and
rjags objectsMCMCplot now takes guide_axis argument. If
TRUE, a second axis (x-axis if HORIZ = TRUE,
y-axis if HORIZ = FALSE) is plotted to help interpret
values on plot.MCMCchains now assigns arbitrary names to columns
(parameters) when input type is matrix (along with a warning that it is
doing so)MCMCtrace now modifies the layout of trace plots when
< 6 plots are generatedMCMCpstr fix bug that prevented scalars when type =
‘chains’MCMCsummary digits argument is now NULL by
default (all computed digits are returned upon default - any rounding
must be explicitly specified)MCMCsummary digits argument uses
signif rather than round for rounding (in
other words, digits specifies number of significant digits
rather than number of decimal places)MCMCsummary now takes round argument to
round output to specified number of decimal placesMCMCpstr no longer has the option to restrict the
number of digits output (returns all digits)MCMCpstr now takes type as an argument.
When type = 'summary' (default), values calculated with the
func argument (default mean) are returned.
When type = 'chains', posterior chains are returned while
preserving the parameter structure. Posterior samples are stored in the
last dimension of the array for each element of the output list. In this
way vector parameters are output in matrix format, matrix parameters are
output in three dimension array format (within each element of the
output list - one parameter for each list element).MCMCpstr now accepts output greater than length 1 from
argument func. If output is greater than length 1, function
output are stored in the last dimension of the array (within each
element of the output list - one parameter for each list element).MCMCchains now takes chain_num as an
argument. When specified, single posterior chains can be output for a
particular parameter of interest. Useful for determining the last value
in an MCMC chain for each parameter (to be used as initial values for a
subsequent model run).excl and params arguments instead of an
error. This means that output will be returned even when values
specified do not exist. Change was made to MCMCchains code
but impacts all functions.MCMCtrace now takes post_zm as an
argument. When post_zm = FALSE, x- and y-limits of density
plots are scaled so that both the prior and posterior can be visualized
on a plot (rather than zoomed on the posterior).MCMCpstr fix bug to institute rounding when parameters
as scalars or vectorsMCMCtrace now takes PPO_out as an
argument. When PPO_out = TRUE, the percent overlap between
prior and posterior for each paras will be returned as a
data.frame.MCMCsummarystanfit
objectsMCMCtrace now takes open_pdf as an
argument. When open_pdf = FALSE, the generated pdf will not
be opened in a viewer automatically.MCMCtrace now takes gvals as an argument.
When simulated data are used to fit a model, the generating values used
to simulate the data (true parameter values) will be plotted as vertical
lines on the density plots.MCMCplot ref_ovl argument now defaults to
FALSE (one color is plotting for all parameter estimates)MCMC_data (example data) so that it’s smaller
(only 5k iterations and two parameters)MCMCplot when many parameters
are plotted and large fig dimensions are usedMCMCplot when
horiz = FALSE and large numbers of parameters are
plottedMCMCtraceMCMCtrace now takes matrix input (as with the other
functions). One chain is assumed when matrix input is used.jags.parallel function in the R2jags
package.ISB = FALSE, params
argument now takes the form of regular expressionsMCMCtrace no longer open up external
programs (pdf viewer) per CRAN policyMCMCpstr function now added. Function returns summary
output for a specified function while preserving structure of parameters
(i.e., scalar, vector, matrix, array).MCMCtrace now takes a priors argument to
visualize prior/posterior overlap. If specified, the prior (user
specified as this information is not contained within the MCMC output)
for a specified parameter is plotted on the same plot as the posterior
output. Percent overlap between posterior and prior is also calculated
and displayed.MCMCchains that caused incorrect
alphabetization of parameter names when output from R2jags was
used.MCMCsummary greatly speed up calculation of Rhat values
for objects with large numbers of parametersMCMCchains now takes the argument
mcmc.list. If specified, mcmc.list object
returned rather than a matrix.MCMCsummary that displayed the same result
twice when selecting only a single output parameter of interestMCMCplot that displayed the axis label too
close to tick labels when horiz = FALSE and tick labels
were very longMCMCsummary Rhat values always round to 2 digitsMCMCsummary output from func argument
rounded to specified digits for rest of outputMCMCsummary now takes a func_name
argument. If specified, column displaying output from func
will be labeled with this name. If not specified, column will be labeled
‘func’.MCMCplot change argument x_axis_text_sz
and x_tick_text_sz to axis_text_sz and
tick_text_sz respectivelyISB (Ignore Square Bracket) has
now been added. By default ISB = TRUE - params
and excl match exactly to parameter names by default
(ignoring square brackets). When ISB = FALSE, square
brackets will not be ignored, and will match on partial names (as when
using grep). This applies to all functions.MCMCsummary now takes a func argument. If
a function is specified, it will be evaluated for all specified
parameters and specified in the MCMCsummary output.MCMCsummary speed greatly increased. Parameters of
interest are now sorted before calculations are made. Rhat values are no
longer masked, but rather not calculated when Rhat = FALSE.
These changes result in dramatic speed ups for large objects.MCMCsummary bug fixed that caused function to fail when
only one chain was runMCMCsummary standard deviation added to summary output
for each parameter.MCMCsummary number of effective samples added to
summary output for each parameter. Default is n.eff = FALSE
(metric will not be calculated or displayed).MCMCtrace default is now to write trace plots to pdf.
Default number of iterations changed to 5000 from 2000.MCMCplot y-axis labels now vertical when
horiz = FALSE to improve readability.MCMCplot bug that resulted in poor plot dimension
choices in some circumstances now fixed.MCMC_data now contains three chains with 6000
iterations each.jags.samples function in the coda
package. coda.samples should be used instead.MCMCplot which incorrectly shaded parameter
estimates when plotted verticallyMCMCsummary now displays estimates for deviance with
MCMC output fits with R2jagsMCMCsummary to do with Cholesky
decomposition and calculating Rhat.MCMCsummary for certain object
types.MCMCplot labels now start at top and go down (more
intuitive).MCMCtrace now plots only the last 2000 iterations of
the posterior chains by default. The argument is now number of
iterations to be plotted, rather than the starting iteration to plot. As
such, MCMCtrace argument iter_st (start
iteration) changed to iter (number of iterations from
end).horiz argument to MCMCplot -
caterpillar plots can now be plotted to run vertically rather than
horizontally. Parameters are plotted left to right when plotted
vertically.MCMCplot - axes lines
only goes to the end of ticks now.These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.
Health stats visible at Monitor.