umx_make("release") no longer prints the
devtools release() deprecation. The CRAN upload name is one
string, Timothy C. Bates."acov").umxACE_DE double-entry ID: overwrite Paras Mehta -
avoids forcing continuous trait variance to 1. All DE pairs release
binary Vtot==1, share free cont/cens means;
fixCensorThresholds fixes thresh at known cut or leaves
thresh free (cut estimated in data units).umxMI runs on multi-group models, and notes/skips
un-identified changes.umxMI Vignette inc. multi-group example.xmu_make_mxData subsets modern summary WLS
observedStats (useWeight / asymCov).xmu_gsem_subset_covstrucumxGSEM_std function to standardize GSEM S and V
matrices (Delta method) and scale intercepts.umxSummary for covstruc/mxData Nice
for understanding the new covstruc format from built in Psych_LDSCmxRAM now emits topologically sorted models
automatically (great for optimizer, esp with hard models!)umxGSEM_label_ldsc() labels GenomicSEM-style lists
(S/V/I/N) with
OpenMx residual names (var_* / poly_*_*).umxGSEM() fits one model to LDSC S/V via modern
OpenMx.umxGSEM()Accepts lavaan/umx strings OR
umxRAM/mxModel structures.umxGSEM_sumstats() allele-align GWAS files to a
1000G ref (baseDir for basenames).umxGSEM_GWAS() per-SNP expand of S/V, DWLS fit,
table of factor~SNP effects.umxGSEM_sim_snps() fake
beta.*/se.* table for
umxGSEM_GWAS() smoke (not real GWAS).umxACE_DE double-entry ACE model for censored
data.umxLISREL make LISREL Models!plot supports MxLISRELModelsxmu_standardize_LISREL standardize LISREL
models.S,
I, V, and N.umxGSEM_dl_tutorial_files() (not a data()
object; munge/GWAS smoke only).umx_scale_reliabilities accumulates the
reliabilities computed across calls to umx_score_scale() as
markdown table.umxMeasurementInvariance runs
configural/metric/scalar/strict multi-group CFA sequences, returns
comparison tablesumxGxE handles ordinal DVsumxPlotCP Show CIs (or SEs via mxSE)umxPlotIP Show CIs (or SEs via mxSE)umxCompare: constraints affecting df now
noted for user!umx_make_double_entry_data stores
attr(*, "umxDoubleEntry") (cut/side/fixable); left-censor
levels are c("censored","observed") with censored state
assigned by name (not levels[2]).
Free-threshold models may see thresh sign/direction change; default
still free thresh.umx_tutorials to let users list and open package
vignettes interactively.umxCompare / umxSummary WLS +
Satorra-Bentler 2010 / robust metrics notes for GSEM.umxRun shows fit change for run models.
messages distinguish WLS Chi-Square vs -2LL.umxStart values are MUCH better.umxParan takes cov input (and n)umx_lower2full preserves names, pads
non-square input.umx_complete_dollar makes console
autocomplete MUCH BETTER.umxACE_DE: Added doubleEntrySuffix parameter
and improve detection logicplot.umx_GWAS umxGSEM Expand and
clarify docs and examples for GWAS plottingvignette WLS example code, notes on
misspecification, formatting).umxRAM now adds the fixes means for binary
vars at 0 when adding these for a wayward user doing WLS with raw
data.umxRAM now shows the user the correct mxPath
to add for WLS means.umx_make("dev") will install development
version of umx for user.umx_aggregate support multiple DVs and
output=‘string’ for easier paste into procedure.umxRun passes intervals argument through to
xmu_safe_run_summaryumxSummary now robust to models that lack
standard fit indices.umxSummarizeTwinData add a sex parameter,
print mean age with SD, M/F counts, 95% CIs for correlations.umxDoC Better mean startslibs now works with symbols as well as quoted
stringsumx_msg Refactored to improve debugging
output and robustnessumx_makeRefactored to normalize package path,
tidy function arguments, and use pkgPath throughout.umxPlotPredict now locates “r= xx” string
automatically.umxAPA has tabular output for lm etc.xmu_topo_sort documented (used to aid umxRAM
optimization)umx_cor can take a formulafin_tax_FIF() applies FDR to opening equity
plus assets bought with the loan. Neutral loan is \(L/E = ft / (r - i(1-t) - ft)\). Prints
loan/assets; plot LTV is loan/opening.umxGSEM_ldsc() calls OpenMx imxLDSC
on munged .sumstats.gz and returns a labeled covstruc
(var_* / poly_*_*) for [umxGSEM()].umxGSEM_dl_tutorial_files()
w_hm3.snplist,
eur_w_ld_chr/) into
tools::R_user_dir("umx", "data")/GSEM_tutorial.umxACE_GLM Cholesky on
mxFitFunctionGLM (nVar-general; MZ Laplace
dimension 4 * nVar). 3-trait Poisson recovers
a diagonals / a_r2c1 on eta.
plot() / umxPlotACE_GLM is the compact a/c/e
diagram (printed e is chol(ee’+D)).
e diagonal fixed at 1 (scale).
trials= / per-row n columns are not
implemented (Bernoulli only; no beta-binomial).mxComputeOnce('fitfunction', 'information', 'meat') then
mxComputeStandardError().lbound = 1e-4.phi = 1/theta) so it does not collide with e.
Documented as not for ACE (trial-level RT / first-passage; twin ACE is
one phenotype per person).e Cholesky kept (unique
covariance).umxSummaryACE_GLM uses E = ee' + D
(D = 1 probit, pi^2/3 logit).umxSummaryACE_GLM bivariate prints rA,
rC, rE = cov2cor(E_total) (unique correlation
diluted by leftover; rA is not).umxCompare no longer asks
umx_is_MxModel() whether its comparison list is one model.
Avoids “set listOK = TRUE” printumxCompare / xmu_compare_WLS use
GSEM DWLS chi-square only for MxModelGSEM.
numObs > 50000 no longer routes ordinary WLS onto the
genomic track.umxGSEM_munge maps columns per file (tutorial
BIP snpid/pval vs SCZ
SNP/P). \dontrun example uses
umxGSEM_dl_tutorial_files().umx_is_GSEM() is TRUE only for class
MxModelGSEM. Ordinary RAM models are no longer treated as
GSEM via MxRAMObjective.tmx_show.MxModelGSEM defaults to A,
residual S, data.S, data.V (LDSC
cov / sampling cov in observedStats). Bare "V"
is not a RAM matrix.umxGSEM string parse uses a PD identity named by
traits, not raw S. WLS still gets the real (possibly
smoothed) S/V.umxSummary/plot use cont factor
cols not 1:nKeep;umxSuperModel handles more complex nested
models.umx_tail() prints the n most extreme
rows on a column. what is high,
low, both, or abs (furthest from
the median). Markdown by default; report = "html" opens the
[umx_print()] table.gh_message_search() lists commits whose message
matches a regex, newest first, with the GitHub commit page.
open = 1 shows that row in the browser. Part of full git
support.umx_yj_wide_twin_data() uses
car::powerTransform / yjPower (Yeo-Johnson).
Dropped Suggests bestNormalize (and its large dependency
tree).tests/testthat.Rinst/developer/GenomicSEM (toy SNP subsets,
developer tests) is .Rbuildignored; tutorial files come
from umxGSEM_dl_tutorial_files(), not the CRAN
tarball.mxFamily via get0 in
umxACE_GLM / umxRAM_GLM (GenomicMx-only).
umxPlot imports ggplot2 .data.umxModify now validates labels and suggests fixes
from similar names if not found. Prevents silent failures when calling
omxSetParameters.umxRAM Validate dot items & fix loop with
clearer error messages.umx_paste_names usage to use suffixes.umx_log_wide_twin_dataumx_yj_wide_twin_datafin_option teaching about optionsfin_StockCAGR compute CAGR from stock
feedfin_carryCost compute the rent vs. invest
cost of shelter via carry cost conceptumx_wide4lmer make data wide without having to
understand the arcane reshape symbologyumxPlotPredict can do R^2 or r, italic
formatbucks knows about
umx_set_dollar_symbol.umxSummarizeTwinData wraps M/F reporting in a
try-catch.umxSummarizeTwinData prints non-NA data for
each twin group and each variable.xmu_check_variance conditionally reports
too-small data variance.umx_score_scale handles INF in max mode.umxDiffMZ clean up.umxGxEbiv checks selDefs for
user.fin_expected for expected return based on current
and fair value and cost of capital.fin_valuation computes net present value, gets
ticker, verbose mode.fin_expected gets verbose mode, new
examplesfin_interest off by 1 error for “final”
mode.umxPlotPredict to plot predicted outcomes nicely
with an R2twin models now know about SEstyle = TRUE for mxSE
SE reporting.umxSummary Has improved notes: showing what
column names abbreviations are (for publication)umx_show control over parameter colors,
examplesumx_wide2long much more capableumx_score_scale correctAnswer mode now works
with NAsumx_time can return current time.umxCP standardized model reporting.umxPlotFun log scales.prolific_anonymize alsoDrop naming.xmu_get_CI under the hood improvements.umxConfint edge case handling.umxAPA gains html report for lm modelsumxAPA pretty printing enhanced.umx_scale_wide_twin accepts sibs #253 from
lf-araujo/patch-6umxBrownie makes a rare steakumxSummarizeTwinData reports complete pairs
for rMZ and rDZumx_as.numeric explicit drop = TRUEumxReduce ACE can skip ADE modelumxReduce output better formattedumx_check_parallel gains set optimizer, and
row parallelumxEFA error reporting is more robustumx_scale betterumx_scale clarify polychoric and pairwise
functionalityumxRAM Can include definition variables! h/t @lf-araujoumxRAM example a model with a def var on
meansumxMR can standardize manifests.prolific_read_demog outputs a ready to paste
demographics string.prolific_anonymize can delete more
columns.umxReduce.MxModelACE now aware of
umx_set_auto_run #241 @lf-araujoexpression instead of
using uXXXXumxMatrixFree undocumented argument ‘byrow’umxMRDoC sketch mode fixedfin_ticker open a stock in the browser.umxTwinMaker can manage definition variables
(by @lf-araujo)umxSummary suppress duplicate pathsumxGxE_window handles target range without error
once more (h/t Marla Kürten)umxMRDoC byrow misspellingumxMRDoC Direction of causation with twins! (by
@lf-araujo)umxMatrixFree to help in simulations (by @lf-araujo)prolific_anonymize() remove IDs and other
identifying columns: Ideal when sharing data.ggAddR helper to add R/beta to a graphumx_strings2numeric helper function to recode
strings to numbersRMSEA()umxAPA gains cols to allow input
checking in place of Rs…umx_strings2numeric example, handle 1 col
dfprolific_read_demog handle
“DATA_EXPIRED”umx_score_scale reliability works with
correctAnswersumx_is_numeric works with tibble input h/t
@psychelzhlibs can for package update using
force.update = TRUEumxAPA cor/means output data.frame inputumxModify works with new R null behaviorumxAPA (and others) report = “html” works
properly.umx_score_scale omega refsumx_read_prolific_demog ->
prolific_read_demogumx_merge_randomized_columns wide-2-mingled
(rather than wide2long) to unify data from people assigned to different
conditions.umxDiscTwin Discordant twin test for causal
effects.umxDiscTwin Graphical output for publication.umxPlot ease customized/journal-close, e.g. y~x,
-> x y plot, with r^2.umxAPA t-test reporting.umx_score_scale gains error and
proportion.umx_score_scale report only omega total when
1 factor extracted.umx_read_prolific_demog gains
functionality.umxMR: gained `tryHard’umxGxEwindow: gained tryHard,
smarter print control.umxGxEwindow: removes un-needed columns.umxGxE checks that selDefs are
numeric (closes #196, h/t @salvatoj)umxRun: gained summary
control.umx_read_prolific_demog more control, better
defaults, and more reliable summary table.umx_score_scale cleaner output for 1 factor
case with omega.xmu_twin_print_means now takes >1 covariate
h/t @Eren Kafadar
#201xmu_safe_run_summary now takes refModels = FALSE
h/t @lf-araujo
#200umxACEcov plot and summary working (needed to
filter covs from DVs).xmu_twin_print_means handles more than 1
covariate and improves labeling.umx_score_scale handles NA correctly.umxPath stops when v*m* is
used with firstAt or fixedAt.umxAlgebra warns that it isn’t
functional.umx_print call fixed where was being called with
file.umxCP help .svg figure instead of
`.png`` (smaller).umxMR Remove unsupported ‘weights’
parameter.umxDoC can take data (#ht @lf-araujo )umxDoC plot Rd figure includes
specifics.umxPower search plot with fixed N has effect
size on the x axis.umxGxE plot legend location works for
ggplot.umx_score_scale gains omega reliability.umx_score_scale bug in handling new
mapstrings.umxDoC plot.omxLocateParameters call was breaking
umxSummary.xmu_safe_run_summary supports
intervals parameter.umxSummary works with umxSuperModel
now that paths have types.umx_make_twinData works with custom namesumx_make_twinData no longer shifts means now
that mean and sd parameters are available.umxGetModel,
umxGetLatents, and `umxGetManifestsumxPower can plot power search space! (beta!)noNAs() helper returning complete rows of a
data.frame.umxThresholdMatrix can handle one value
columnsumxPath gains v0m.umxPower parameters ==1 for power search with
fixed numxPower refpower.ACE.test can run ADE modelpower.ACE.test Clarify which of power,
effect, and n are fixed or being estimated.umxReduce.ACE gains tryHard (compute
ACE/ADE/CE/AE, outputs a table an also AIC-based probability weights for
which is the preferable model)umxReduce tests familiality (E model)umxReduce(ACE) includes the parameters of the
reduced models. #ht @nick martinumxSummarizeTwinData takes an age column and
notes when it can’t be foundumx_score_scale can score items with
character labelsumx_score_scale gains the ability to report
Cronbach alphaumx_score_scale changed default
na.rm = TRUE.umxAPA gains OR output for logit modelstmx_show handles matricesumx_rename better names for deprecated
parametersumxRun has tryHard (not a very useful
function)umxCompare can silently return a data.frame
for further modificationumxCP warning to not use
correlatedACEumxAPA (invisibly) returns the standardized
model for lmbucks more control for formatting moneylibs now install.packages() if not foundumx_make_TwinData shifts data to be positive
if it not being scaledgrViz() in place of print(DiagrammR())`
(should work better with knitr)fin_interest more memo/type-able parameter name
for inflationumxRun defaults to tryHard= “yes”umxPower supports multiple path drop.FishersMethod Rd refumxCP ref Martin, N. G., & Eaves, L. J.
(1977). The Genetical Analysis of Covariance Structure.
Heredity, 38, 79-95.deg2rad Rd better in htmlumxRAM sketch mode helpumxCP more refs for RdumxACEv and umxACE work with
nSib = 3!umxCP correlatedACE was not being
filtered out of the expectation #hat-tip @micanzach But still working on what the
right choices are here…umxCP defaults to tryHard = “yes”umxRAM deprecated parameter show
after 4 years deprecationtmx_show works for M matrix, closed #163kable and kbl get format passed in
at all callsumx_make_TwinData supports nSib
> 2 for MZ/DZr modeumxAPA more use of cat to print (easier to
read and copy)umx_open_CRAN_page printing if installed
version if foundumxAPA more use of cat to print (easier to
read and copy)fin_interest compute effective interestxmuTwinSuper_NoBinary update for nSib >
2kable formats (pandoc “pipe” and “simple” added,
“pandoc” removed)add_footnote in place of
footnote…umxPath and umxRAM helpxmuTwinUpgradeMeansToCovariateModel can handle
nSib=3xmu_twin_make_def_means_mats_and_alg helperxmu_start improved closes #158umxPlot now detects and plots Pearson-Aitken
selection paths (solid red headless).umxPlotFun plot functions without hassles. e.g
umxPlotFun("sin(x)")umxModelNames to get model and submodel names from
a model.libs as a short cut for
multiple library() calls
libs("x", "y", c("a", "b"))fin_valuation uses revenue, margin, expenses and
PE to compute a market cap.fin_interest shorter parameter name
dinflate and other improvements.umxACEv Tables now include
markdown-compatible table captions.umxAPA reports R^2 for lm (in
addition to beta & CI[] for each parameter)umxSummary sorts RAM parameters by type
(factor loading etc.) and can filter means and residuals for ease of
use.umxSummary gives desired levels for good TLI
and RMSEA.umxSuperModel checks for duplicate names in
incoming model names and makes them unique.umxCompare Maintain compatibility with new
OpenMx changes to mxCompare (esp. for WLS).umx_read_lower accept names for lower
diag.umxCP correlatedACE leaves the
covariances fixed@0.umxACE added A_std closes
#151umxAPA pretty printing.umxPlotFun check for character input.xmu_summary_RAM_group_parameters sort and filter
RAM model parameters for ease of understanding.umxFactor now used in all examples involving
factor making.plot can output to pdf, png, svg!umx_set_plot_format adds pdf, png, svg plot output
(in addition to DiagrammeR and graphviz)umx_lower.tri helper to get lower.tri values in 1
callumx_set_plot_use_hrbrthemes controlplot.percent plot support for percent change
questionsumxSummary now reports algebra values when
present in RAM models closes #70umxSummary tables have pandoc captions (see
#145)umxCompare table output and plain-English
summary printing (cat)umxCP practical workflows for correlated
ACEfin_compound_interest now outputs a table of
annual returnsumxAPA uses cat to improve usability of
summary dataumxCP Document that correlatedA works on a, c and
e, allowing factors to correlate.umxGxEbiv “lboundACE” @ 0 by defaultumxSimplex exampleumxGxEbiv fix nonfunctional lboundACE and
lboundM (thanks @mcneale)umxGxEbiv selVars order was imposed on
expectation dimnames but didn’t match the cov matrix due to bespoke
nature of expCov algebras…tmx_show display model matrices on the web, with
color-coded free/false, values + roll-over labels.harmonic_mean with support for weights.harmonic_mean Example averaging speeds across
distance.geometric_meanfin_compound_interestfin_percentumxPlotGxE Gains CI ribbons (closes #96)umxPlotGxE Defaults to use ggplot and now
returns the graphs for further decoration (closes #139)umxPlotGxE Can now plot user-specified
moderator ranges (closes #96)umxPlotIP Now relies on matrix cells, not
their labels (closes #61)umxACE Returns group models invisibly #134
@khusmann.plot Now allows “ortho” “compound” and
“polyline” spline control.umx_set_optimization_options Can get and set
“Parallel diagnostics”tmx_show Smart about not printing NULL
matrices.Gained anewlabels` parameter for
the equated paths (rather than one path adopting the label of the
other)umx_score_scale Supports
minManifests for factor scoresumx_open_CRAN_page Can now install a package,
as well as open the CRAN page for the package.umx_rename_file Better feedback about matches
and files checked.umx_long2wide More resilient to zygosity
column type.umxSummarizeTwinData More resilient to name
of age columnplot strip_zero option works for
fixed (@) parametersumx_is_class Handle non-df input.umxCI: Gains regex parameter to
support adding and removing CIs on labels matching a string
pattern.umx_is_class Handles non-df input.umx_set_table_format Defaults to “markdown”
on load.umxPlotIP Wasn’t working with threshold models
(didn’t ignore thresh names for plot).umx_time Was using time units for model 1 for
all models… h/t @khusmannumxPlotACEv Was not displaying half the
variables in an MV model as NA.umx_set_cores No longer use
imxGetNumThreads which is hard-coded to not report number
of threads.tmx_show Has html_font parameter in
help.CatchesmxRefModels` error for
WLS models and politely informs user OpenMx doesn’t support fit stats
for WLS yet.umxThresholdMatrix Changed parameter from
selDVs to fullVarNames to prevent user error.umx_read_prolific_demog to read prolific academic
demographics files.umxRAM Groups in multi-group models now
independent by default (label style “path_group” )umxRAM No longer demand new data if a model is
given as the first parameter. Updates data if provided.umxACE Return group models invisibly #134
@khusmannumxACEv support for covariates enabled.umx_score_scale gains the ability to return a
factor score.umxEFA returns scores by default when
requested (used to require user to also set
return = "loadings" (thanks to user!)umx_r_test handles case of overlapping
variables, i.e., compare r(ab) to r(ac)umxPlotACE uses matrix cells, not labels to
draw the plot (so now labels are arbitrary)umxIP checks names of nFac for each of a, c,
and eumxSimplex default sep is now “_T” instead of
NULL (should do this throughout?)residuals can now suppress 0s - helpful for
large tablesxmu_dot_mat2dot can cope with 1-variable name
listsumxValues -> xmuValues clean
up namespaceumxLabels -> xmuLabels clean
up namespaceparameters: std wasn’t being used.umxEFA examples.umxTwinMaker.umx_score_scale Fix a bad design decision where
I was storing the reversed items in the data returned to the user.
Worked for 1 run, but not repeats…umx_long2wide now looks for zygosity in more
than just twin 1 h/t @nathangillespie for catching this
case.m1 = umxACE(selVars = "wt", selCovs= "age", sep="_T")umxTwinMaker: Make twin models by specifying
just the within-twin paths for one twin
umxSuperModel and runs it!plot() and “std= TRUE” work for any model now!umxDoCp Path-based Direction of causation
model.umxSummaryACE prints a means table!
plot for umxACEv now prints model
name on the diagram.plot for umxACE now uses tips for
printing.FishersMethod: accepts p-values for
meta-analysis in “…” argument.namez handles class(list) objects, returning
$ components.parameters supports standardizing parameters
(closes #92) .power.ACE.test now prints the parameter and
value when testing power (closes #111).umxCompare includes Delta-AIC.umxCP plot prints model name.umxGxE code to detect mismatched
moderatorsumxGxE code to include co-twin moderator in
means modelumxGxE simplified algebraumxIP gains report option.umxIP plot prints model name.umxSummaryGxEbiv working, with a note that
the whole function is beta. h/t @MikeNeale.umxThresholdMatrix re-implemented
l_u_bound - wasn’t working for some time h/t @NathanGillespie.umxThresholdMatrix supports leaving all
thresholds free with method= "allFree" h/t @NathanGillespie.umx_check_names returns TRUE when no names
are entered.umx_check can concatenate extra messages, and
is by default polite with messages.umx_check handles … items betterumx_is_ordered can process mxData as well as
df.umx_make_TwinData has dzAr.umx_make_twin_data_nice can turn numeric zyg
column into human readable “MZMM” etc.umx_print gains option to append to existing
file.umx_print gains option to make tables
sort-able.umx_print gains option to output content in
“both” md and html.umx_print supports report to html.umx_rename now uses “from=, to=” format for
renaming columns.umx_rename was using old/replace instead of
from/to (and so generating a polite note).umx_residualize now prints the betas (and
CIs) of the residualization it performs.umx_score_scale checks input rangesumx_score_scale tell user what the max
response is for reversed items if they leave this blank.umx_score_scale Warns user if there are NAs
when “score” is totals.umx_rename_file: Perl mode is TRUE (allows
lookahead/behind features; default list pattern is NULL.umx_rename_file now ignores file suffix by
default.umx_rot can NA the last value.umx_var upgrades input to df (e.g. if you
feed it a column, it now works).xmu_data_missing can hint= to
user which dataset had missingness.xmu_make_mxData handles dropping rows,
alerting about variance ratios.xmu_make_mxData now handles cov matrix input
(added numObs parameter).umxACE improved tutorial, more tips * IMPROVED
plot for umxACE now prints model name on the
diagram.umxACE covariate examplesumx_explode_twin_names links for
find-abilityumxACE residualize examples matchumxRenameMatrix wrap up the multiple lines of code
needed to modify a matrix in OpenMx.umx_str_chars return selected characters of a
string.umxTwinUpgradeMeansToCovariateModel function to
add def-based covariates to means in umx twin models.xmu_data_missing helper to drop rows with missing
def vars or stop and alert the user to do this.xmu_extract_column to allow mxDataxmu_twin_add_WeightMatrices helper.xmuDefMean helperxmuDefBetas helperxmu_twin_upgrade_selDvs2SelVars helper.xmu_make_TwinSuperModel handles selCovsparallel::detectCores –>
imxGetNumThreads (as OpenMx does).plot for umxTwinMaker models!!xmu_path2twin twin model helper for
umxPath usersxmu_path_regex helper for making twin models for
umxPath usersumx_file_load_pseudo to load and process
non-transmitted genome files for the method we developed in Bates, …
Gillespie, N. A. (2018). The Nature of Nurture: Using a Virtual-Parent
Design to Test Parenting Effects on Children’s Educational Attainment in
Genotyped Families. Twin Res Hum Genet, 21(2), 73-83.
doi:10.1017/thg.2018.11xmu_data_missing helper to drop rows with missing
def vars or stop and alert the user to do thisxmu_make_TwinSuperModel handles selCovsumxFitIndices Many new indices! @bwiernikumxPath(v.m0 = ) now uses values (if
provided) to set starts for varianceumx_check_parallel use all cores (instead of
half) by defaultumxPlotACE variable labels (trimmed off _T1
like umxPlotCP doesumxDoCp bound e/es above zeroumxDoCp start with all variance as
specificpower.ACE.testumxLabel 100x speedup for RAM and Matrix
model labeling!xmuLabel_RAM_Model 100x speedupxmu_safe_run_summary has smarter default for
std (model summary default)xmu_dot_mat2dot now accepts algebra input
begin umxPlot for DoCumx_check handles … items betterxmu_make_mxData handles dropping rows,
alerting about variance ratiospower.ACE.test no longer fails when estimating
power at a fixed numxSummary: Improve Rd fileumxACE covariate examplesumx_explode_twin_names links for
find-abilityumxDoc example but summary and plot TODOumxRAM2Lav export umx to lavaanxmu_safe_run_summary has smarter default for
std (model summary default)umx_score_scale: implemented score = “mean”
and score = “max”umx_score_scale: better documentationumxGxE gains dzAr and dzCr control so ADE is
possible as well as assortative matingxmu_twin_check works with mxData * IMPROVED:
xmu_twin_starts works with mxDataxmu_make_top_twin works with mxDataxmu_twin_check support mxData (was just
data.frame input)umx_long_2_wide: gains some notes on
processing TWINID into famID and twinIDumxSummary handles std=NULL as documented
(don’t show parameter table)umx_nice_data –>
umx_make_twin_data_nice with future improvements planned to
cope with non numeric twin schemes.umx_is_ordered was broken by previous fix to
support the un-tidyverse mucking with dataframes… Now fixedumxRAM supports multi-group via `group =
“column”umx_score_scale now works with binary (0-1)
itemsumx_time gains “lap” (report elapsed time
without restarting timer)umx_time behavior with no input - now prints
the current time and dateumxFactorScores can now return
SEsumxFactorScores allows abbreviated type
namesumx_reorder can operate without verifying
data (ignore precision digits in triangles of cov matrices)residuals can re-order variables for ease of
checking patterns of residualsumxPath(defn= Definition vars now auto-named
if labels not given closes #107umxACEv uses default name “ADEv” (instead of
“ACEv”) if user sets dzCr = .25umxAPA only uses italic Beta when std =
TRUEumxRAM now respects
umx_set_silentumxRAM2Ordinal to
xmuRAM2Ordinalumx_fix_latents and
umx_fix_first_loadings to encourage user to be more
explicit in their modeling.August 2019 R 3.6.1 “Action of the Toes”
This release has major new features in beta including support for
lavaan syntax, AND a over 80 additional improvements clean-out/cleanup
of old functions and parameters that impede getting learning and using
umx: Think of it like ggplot2 version
2.
ALPHA: umxRAM lavaan string syntax support!
ALPHA: umxRAM multi-group models with
group="column"!
BETA: power.ACE.test and examples. (beta)
BETA: umxPower and examples (beta).
BETA: umxRotate Rotate the factor loadings in
umxCP models.
NEW: umxCP and umxIP can take
data and create MZ and DZ
datasets.
NEW: deg2rad rad2deg utility
functions
NEW: FishersMethod To combine p-values.
NEW: oddsratio.
NEW: SE_from_p helper to get SE from b and p, or get
a p from CI.
NEW: umxMendelianRandomization alias to
umxTwoStage.
NEW: umx_nice_data converts your twinData to the
standard format (zyg = zygosity, “_T1” , “_T2” suffixing of twin
variable names).
NEW: umx_select_valid to replace values in one
column with those in another, if first column is NA.
NEW: umx_set_silent preference (option) for other
functions to listen too and choose how much junk to print to console.
Like a global verbose.
NEW: xmu_safe_run_summary can listen to
silent to turn off summary and progress from models - use
when running big simulation loops.
IMPROVED: namez has a better default action (call
names )
IMPROVED: oddsratio teaches about limitations;
supports odds-format as input (closes #102)
IMPROVED: plot can do
pathLabels = "labels" to show the labels for paths
IMPROVED: plot for umxIP supports
means=TRUE
IMPROVED: power.ACE.test now reports searches with
fixed n nicely
IMPROVED: umxAPA now returns its output, rather than
printing it. makes for easier consumption in programmatic uses.
IMPROVED: umxAPA supports cor.test and
t.test
IMPROVED: umxConfint uses smart confidence intervals
(just the free standardized parameters) for umxCP models.
IMPROVED: umxEFA can suppress printing when
umx_set_silent(TRUE) #103
IMPROVED: umxEFA now has an option about reporting
the summary fit statistics of a model (default is FALSE) see
#103
IMPROVED: umxGxE removed border from legend
(obscures plot to no benefit)
IMPROVED: umxGxE supports digits (rounding for
tables)
IMPROVED: umxlav2RAM catch means
IMPROVED: umxPlotCP can show (non-zero) fixed paths
closes #97
IMPROVED: umxReduce.GxE Don’t try and drop means
moderation.
IMPROVED: umxReduceGxE gains a tryHard
option
IMPROVED: umxReduceGxE more rational set of
reductions - means obey principle of marginality.
IMPROVED: umxSummaryGxE prints parameter table and
SEs as well as the interaction plot.
IMPROVED: umx_cor quieter, more informative
IMPROVED: umx_is_class robust to being given a
tibble (closes #101 Might need to re-open if there are more cases of
where toggling drop from TRUE to FALSE interferes with data
types)
IMPROVED: umx_make_raw_from_cov can add names to
generated data
IMPROVED: umx_make_TwinData bivariate parameters
(e.g., aMod) –> bivAmod for more clarity
(closes #78)
IMPROVED: umx_make_TwinData can now simulate D i.e.,
A, C, D & E
IMPROVED: umx_make_TwinData More robust
IMPROVED: umx_move_file supports wildcards (closes
#83)
IMPROVED: umx_rename examples
IMPROVED: umx_rename reorder parameters, call x->
data, deprecate grep in favor of
regex for consistency,
IMPROVED: umx_scale_wide_data gains twins
parameter
IMPROVED: umx_set_data_variance_check: set default
minvar to .1
IMPROVED: umx_set_silent returns old value
IMPROVED: umx_var robustness + support ordinal
variables.
IMPROVED: xmu_dot_move_ranks set min, max or same to
“” to take these ranks out of the diagram - Aids tricky layouts. (closes
#84)
IMPROVED: xmu_make_mxData drop duplicates from
manifests list
IMPROVED: xmu_make_mxData handles manipulating a
1-column mxData input
IMPROVED: xmu_safe_run_summary supports digits
(rounding for tables)
CHANGED: power.ACE.test now uses semantic labels
(“a”, “c” ) and can test a dropped after c. Better feedback for
boundDiag
CHANGED: umx_show ->
tmx_show.
CHANGED: umx_set_optimization_options to
umx_set_mvn_optimization_options for user clarity.
CHANGED: umx_default_option to
xmu_match.arg as programming aids are moved into xmu
space.
CHANGE: umx_aggregate now uses df as default
data.
CHANGED: showEstimates -> show
across all functions: makes learning/usage more consistent.
CHANGED: umxGetParameters doesn’t anchor search
strings to front of label, nor suffix with digit label.
DROPPED: umx_drop_ok - orphan function - just use
umxCompare
DROPPED: umxEval - broken anyway - just use
mxEval
FIX: mxPath v0m0 , v.m0
now use labels (if two provided)
FIXED: umxACE standardizes model in summary
DEPRECATED: The parameter to show standardized parameters is
universally std=TRUE (show = no longer
works).
REMOVE: umxIPold
NAMESPACE clean up to make it easier for users to focus on functions they (rather than devs) will use
HELP: @md links
HELP plot Better explanation of graphing: can still
be improved
HELP umx_as_numeric:better examples
HELP: umxMendelianRandomization nice figure
HELP: NEWS layout improved for R 3.6.0
HELP: Rd doi link, spell-check,
HELP: umxEFA notes it can use formulas,
e.g. umxEFA(~v1+v2+v3, data)
HELP: NEWS updated
HELP: GxEbiv Improved help page
HELP: umxPower gains text and a nice figure
plot splines
= T|F and min= , max = same=
controls to layout your graphs!umxLav2RAM: detect ngroup, and a
umxSuperModel of groups.plot.MxRAMModel plots multiple groups (on separate
pages)!plot gains control of object clustering with
max min and same.plot select whether to allow splines with
spline=T\Fplot opt paths out of layout constraints!plot now labels plots with the group
name.umx_time Handles stop without prior start.
New now option returns current date.uxmSummary the annoying “umxSummary not defined”
error should be fixed!umxSimplexumxLav2RAM: multiple group example in help.umx_set_separator allows Europeans to use ; in
place of comma.xmu_lavaan_process_groupxmu_clean_labelxmu_name_from_lavaan_str helpercitation("umx") Our paper on umx is out!
?umxExamples to see the
example code from Bates, Neale and Maes (2019).umxSexLim multivariate sex-limitation
working.umxSexLim Gained nonlinear, linear, and
homogeneity modesumxSexLim Gained examples.umxRAM supports WLS, UWLS, DWLS models!umxACE supports WLS, UWLS, DWLS models!umxPlotCP Supports CIs on the diagram!umx_make_top_twin_models increases the robustness
and improve-ability of all twin scripts.xmu_dot_define_shapes builds the latents and
manifest shape definitions (should be xmu_)xmu_check_variance checks data for minVar (default
> 0.1) and comparable scale for variables (maxVarRatio default = 500)
umx_set_data_variance_check get/set variance
tolerance.umx is checked on travisplot allows user control over splines, as well as
max, min, same positioning.xmu_dot_move_ranks.xmu_dot_rank_string.namez options for collapse get easier to
short-cut names.data(Fischbein_wt) weight of 66 females record
over six 6-month intervals (from Fischbein (1977) weight data).umxRAM2 Beginning support for lavaan syntax
strings.type (WLS) added to twin models…umxValues to work more reliably with data with
no means.umx_rename Didn’t give correct values when
test = TRUE for non-grep searches (drop NAs)umxSexLim match.arg tryHardselVars vs. selDVs across different
functions (search for “# New-style build-block” to see these
changesumxSuperModel can take a list of models.tryHard - allow “yes” for default mxTryHard
across all twin and RAM models.umxPlotCP revamped - now not dependent on
labels! (old function now umxPlotCPold)xmu_model_needs_meansumxDescribeDataWLS to check whether the form
of WLS requested will have means of not.umx_make: use check_win_devel; set working
director for release; spelling checkxmu_dot_mat2dot add model,
toLabel and fromLabel + fromType
& toTypexmu_dot_mat2dot examplexmu_safe_run_summary now copes with warnings
in models, summary, compare etc. (as it should have)umxModify now supports find and replace
labels with grep (not just find with grep and set free/value)xmu_assemble_twin_supermodel no longer take
bVector (uses existence of weights instead)umxModify now obeys verbose
optionumx_time reports % change between model
timesumIP helpumx_make now supports quick-make as
defaulttmx_show printing.umx_score_scale help and examples.umxModify now supports find and replace
labels with grep (not just find with grep and set free/value).umxModify now obeys verbose.umxValues and elsewhere more robust to new
data types.umx_read_lower examples and input
checking.xmu_cell_is_on now has lower and upper
triangle with or without diagonal includedxmu_dot_mat2dot now can get parameter values
with CIsumxRAM feedback when no data providedxmu_make_top_twin lose “models” for shorter
name.tmx_show (was umx_show - it is
mostly a teaching functionxmu_dot_define_shapesxmu_dot_rankumx_mat2dot –>
xmu_dot_mat2dotumx_APA_model_CI –>
xmu_get_CIumx_add_variances old function not used and
not very safe.thresholds option from several
functions (always doing deviation-based, WLS isn’t thresholds, and
left-censored will be implemented differently)umxSummarizeTwinData to create summary tables for
papers using twin Data.umxRAM: support definition variables to some
extent in umxRAM.umxMatrix advice user when they specify
umxMatrix("me", 1,1).umxModify nicer free-parameter report as
default output with no changes requested (calls parameters).umxRAM preserve definition variables in
data.plot strip_zero more reliable.plot dropped deprecated showMeans and
showFixed from plot after 3 years (use means= and
fixed = ).umx_scale_wide_twin_data dropped support
for deprecated suffix parameter.suffix as a synonym for
sep removed after 3-years of deprecation warnings.umxSummaryfor ACEv models: CI now
works - (was ‘a’ not ‘A’).xmu_safe_run_summary Runs a model safely,
optionally summarizing and comparing it to a base model.tryHard as a
parameter!!autoRun option in more places.plot functions now have
strip_zero option (default = strip leading zeroes from
parameter estimates)umx_standardize_ACEv now working!xmu_safe_run_summary so even bad
models are returned for the user to diagnose them.umxRAM now labels paths when no data are
provided (simulation of sketch modes).umx_score_scale can cope with a single
item.xmu_safe_run_summary don’t error on code red
non-run models.xmu_twin_check tells user how to rename
selVars and sep when sep is not provided, but is enforced.umx_as_numeric allows user to select which
columns to convert.umxSimplex start values now flexible and
robust (was hard coded for 4-times).xmu_simplex_corner Takes a full values list
(not just 1).umxAPA uses .Last.value as
default input if none provided.umx_make supports
what = "rhub".umxlong2wide now allows user to retain only
desired twin IDs.xmu_mean_var_starts uses “expMean_” as the name
for means cells (was “mean”).xmu_safe_run_summary don’t return bVector
(already available)umxACE help diagrams improved with clearer
ACE_matrix picture.umxRAM help gained an example of sketch-mode and
diagram.umx_scale help file improvedxmu_make_mxData function to make mxData functions
out of dataframes (and also drop variables from cov/cor dataframes.plots.umxCompare gains ability to output Weight AIC
conditional model probabilities! h/t @mNivardumxReduce returns AIC weight-based
probability of being best model.umxReduce.MxModelGxE returns best model
(invisibly).umxReduce for GxE, don’t run no-A and no-C
models with moderation in place.namez given a list of models will return the
names of each.namez allow global replace.umx_lower2full can now take a
data.frame.umx_aggregate can now open tables in a web
browser.install.OpenMx can now install a package file
if selected in the Finder (MacOS only)install.OpenMx now works with windows.umx_print more robust to non table
input.xmu_make_mxdata more robust to non dataframe
input.umx_apply robustness.umx_simplex_corner can take a numeric to set
matrix size.umx_long2wide warn if twinID (order) has too
many levels; improved help; @mdumxAPA test parameter had no effect.umxSexLim now works for univariate tests (thanks
to Michael Zakharin for reporting!)umxSummary Corrected Chi^2 df for RAM
models.umxCP handles ordinal, continuous and binary data
umxCPoldumxCPplot re-written to allow arbitrary labels and
correlated factors.umx_set_optimization_options function to set
“mvnRelEps” tolerances which impact, especially, ordinal model.umx_is_class reports classes if not given a
set of classes to checkumx_scale(verbose=TRUE) now lists the vars it
didn’t touch as well as those it didn’t (also formatted better)umx_make_TwinData can set mean and SD of
dataumx_simplex_corner can take a numeric to set
matrix sizeumx_standardize_CP was ignoring existing
CIsumxAlgebra This appears broken in some
circumstances?umx_polychoric,
umx_polypairwise, and umx_polytriowiseumxSimplextwin model!
umxSummary for simplexiqdatlongitudinal IQ twin data for simplex
modelingnamez alias for the ever-useful
umx_names function.umxAlgebra (just allows name first).install.OpenMx URLs for NPSOL and travis builds of
OpenMxumxSummary.MxModel.ACE instead of
umxSummary.umxSummary.MxModelACE # i.e. delete
the last period in the name… Sorry :-(parameters now returns helpful near-match
names as suggestions on not finding actual matches.parameters now handles vectors of regular
selectionsumx_print now handles vectors.umxConfint “smart” feature to request only
free algebra cells for models I understand (umxCP in the first
instance)namez can handle a model summary (accesses
names() of the summary)umxGxE added option to control plot
colorsumxPlotCP can handle >9 manifests issueumxGxE example improvedxmu_simplex_corner helper function for
simplex modeling.umx_make_sql improvedumxIP nFac now allows specifying different numbers
of a, c, and e factors!umxACEv plot works (shows correlated
factors).umxACEv standardize mostly functional.umxEFA can now take a formula to select
variables.umxIP and CP examples improved (3-variables,
GFF data).umxIP reporting of as,
cs, es simplified into compact stack.umxCP reporting of as,
cs, es simplified into compact stack.umxPlotCP labeling of variables improved
(catching more _Tn).umxEFA reports factor correlation matrix
(thanks @ConorDolan)!umxEFA can return a loadings object.umx_makenot supports spelling check.umxACEv diagonal now unbounded by default
(which is it’s whole purpose)umxACEv got major improvements to help
file.umxSummaryCP got ability to print to
html.plot made more robust to unexpected labels
(now that umxModify can write newlabels)!umx_names Added GFF examples.umxSummaryIP was assuming nFac = 1umxSummary works with multi-group models!
umxSuperModel)umxModify supports multiple find strings and
replace strings!umxAPA supports glm, more robust
input checking.umxEquate allows vectors of labels.umxGetParameters has better help.umxHetCor docs now note is will return a pos-def
version if given a cov matrix.umxAPA now has t-test and glm
examplesumx_install_OpenMx (use
install.OpenMx)umxSexLim multivariate twin sex-limitation.umxGxE_biv bivariate Gene-environment interaction
model.umx_make_TwinData Can now make bivariate
moderated twin data.umxGroup function as part of
measurement invariance support.umx_stack Slightly more powerful version of base
stack.umx_array_shift Shifts an item off the beginning
of a list.umxRAM can take lists of paths as input
(contributed by @bwiernik).umxModify can write newlabels.umxAPA can back-out an SE if given b and
CI.umxReRun (use umxModify)suffix parameter (use
sep instead)umx_parameters(model, "below", .1)tmx_genotypic_effect Part of a suite of teaching
functions for biometric genetics!?umxRAM has a sketch mode. just set
data to a list of manifest variable names.umxSuperModel function to automate multiple-group
model creation.?umxACEv Variance component version of ACE.
(beta)?umxACEcov_fixed Beta1: Handles main effects of
covariates in the means of continuous variables.umxModify can now equate parameters (set master =
), update is slave set of labels.
optimiser parameter for umxACE and other twin
models: Set the optimizer in your model code!umx_is_numeric boolean check if variables in a
dataframe are numeric.umx_is_class boolean check if variables in a
dataframe match a desired class.umx_twin_check internal function for common input
checks.umx_set_plot_file_suffix(c("gv", "dot") function
(to control the filename used for figures)/umxACE was broken with large cov inputs h/t
@NathanGillespie.umxAggregate now actually supports user
functions as input…umxACEv variance components
methodumxPath(Cholesky = ) method: supports labels,
bounds, lbound at 0.umxPath(Cholesky = ) method: return one
statement instead of a list for clarity.umxACE and umxReduce.ACE run
intervalsumxReduce works better with
umxACE models.umxEFA Supports
minManifests.umxEFA Works better when returning a single
factor score.umxParameters now supports
digits.umxSummary “inline” reporting now includes
AIC.umx_aggregate works with factor input/
umx_aggregate(sex_T1 ~ zyg_2grp, data = x)umx_long2wide can passalong
variables.umx_wide2long much more powerfulumx_msg supports dataframesumxACE gains ability to set plot format:
format = "graphviz" or "DiagrammeR"umx_set_auto_plot take TRUE/FALSE as
input.umx_names: added a find and replace option
for namesumxRAM in “sketch mode” now plots models
automatically.install.OpenMx from University of Virginia,
travis, or open travis build page.umx_cor automatically drops non-numeric
columns.umxCI now supports setting interval and one-
or two-sided type (h/t @Conor Dolan).R2HTML to
xtableumxParameters now supports non-run models?umxAPA can format lots of things for
you: from dataframes to p-values - try it out! Let me know what you’d
like.?umx_parameters function, report parameter
estimates, filtering by name and value!
umxParametersumx_parameters(model, "below", .1, "loading")?umx_long2wide: Merges long data on famID, for an
unlimited number of individuals in a family (twinIDs).?umx_wide2long: Takes a wide df (currently limited
to 2 per family), & returns a long-format version.umx_MakeTwinData.xtable
(R2HTML is abandoned, strips decimals from AIC, etc.)umx_set_auto_plot(FALSE).forEach option in umxPath. This
is a one.headed version of “unique.pairs”.autoRun.umxGxE plot colors now use the
universal ACE -> RGB scheme.umxACEcov Now tries to detect violations: Age
and Sex are good examples.umxACE.umxACEcov!umxACEcov;umx_check_parallel (1 and n-1);umx_check_parallelumxReduce does a much better job reporting
automated umxGxE model reduction.umxAPA handles lme models.umxPrint error messages.logic.MxModel now has an Rd page.confint. OpenMx now supports this, so I dropped
it from umx (added alias to umxConfint).anova. OpenMx now supports, this so I dropped
it from umx.umx_fix_first_loadings now skips latents with
fixed variance.umxVersion (returns mxVersion +
information on umx).umxACE start values much better for
univariate models.umx_make_TwinData
varNamenThresh to threshold the created variables, and return
as mxFactorsempirical parameter of mvrnormnDZPairs to get equal numbersumx_make_TwinData uses variance
input. This was ambiguous previously.install.OpenMx can install UVA parallel, travis
latest (on mac) or opening the travis list.
install.OpenMx had a broken URL, now
corrected.umxRAM can now be used in “sketch” mode, to plot
demos without data: just add the list of manifests as a character string
to data.
splus m1 <- umxRAM("test", data = paste0("itemC", 1:4), umxPath(unique.bivariate = paste0("itemC", 1:4)), umxPath(var="X") ) plot(m1)umx_cont_2_quantiles now allows returning cut
points, better examples, level_names
umx_cont_2_quantiles lowest threshold was
emptyumxAPA now reports mode for factor dataumxLabel to rename the returned modelsep is now the preferred separator (synonym
for suffix in umxACE, umxCP, umxIP)umxPath docsumx_cov_diag to umx_varumx_write_to_clipboardumx_r_testumxCP example for improved compatibility with
OpenMx 2.7umxPathconfint()umxAPA table outputumx_makeumxPath(a, forms= , arrows = 2) bugumxRAM returns invisiblyumx_aggregate can summarize multiple DVs in a
tableumx_aggregate can return a formatted table
(kable = TRUE)umxAPA returns lower-triangle of data.frame
correlations (by default)umxSE now included in OpenMx 2.7.0!umx_set_plot_format uses silent = TRUE
internally where neededumx_make_TwinData fix bug in DZ moderator code
and A+C computationumx_set functions now show legal options by
defaultumx_set functions have silent optionumxFitIndices RMR and SRMR (h/t Brenton
Wiernik!)umxMatrix with default labels and name as 1st
parameterumxPath(defn = "def", labels = "data.age")umxEquate supports autoRun and compareumx_names can, as a convenience, take string
vectorsumxGetOpenMx aliased to
install.OpenMx or umx_install_OpenMxumxFitIndices additional fit-indices and now
automatically computes reference models if needed.umxRAM remove_unused_manifests no longer
updates manifestVars + leaves variable in data by default.plot “showFixed” deprecated in favor of easier
to type “fixed”umxPath allows unique.pairs
connectionumxRAM handles suffix labels
umxSummary to show parameters from first
model in multi-group
umx_show_options to
umx_get_options
umx_check_parallel typoplot calls from summaryumxRAM can now build and run Joint Continuous
Ordinal models!!umx_show_options function to help users learn the
options availableumxACE can now implement a form of
left-censoring.umxThresholdMatrix can now implement a form of
left-censoring.umxRAM & umxRAM2Ordinal can
turn off reference models computation with
refModels = FALSE;umx_show can show thresholdsumx_is_MxModel can take a list:
listOK optionumx_check_parallel options for row-wise
parallel and number of subjectsumx_time and umx_check_parallel to
work with new S4 list deprecationumxTwoStage shorter run-time on 2-stage least
squaresumxTwoStage FIML-based Instrumental variable and
Mendelian Randomization (beta)umx_make_MR_data to simulate SNP-based Mendelian
randomization data (beta)umx_set_condensed_slots(TRUE|FALSE): Get and set
matrix compressionumx_get_OpenMx (download OpenMP/parallel version
of OpenMx with NPSOL)umx_cont_2_quantiles(x, nlevels = )umx_factor alias to umxFactorplot(resid = "none") was not suppressing
plotting residualsumxFactor now handles dataframes and allows
non-ordered outputumx_CI more flexible, now allows removing CIs
umx_time handling of no input argumentsumx_pb_note better feedback and setupumxModify - if free is a string, it will be
used as input and the found paths will be freed.umxModify - show summary as well when
comparison = TRUEplot defaults for std and
showFixed changed h/t @MikeNealeumx_scale_wide_twin_data argument names for
consistency: “suffix” “data”umx_residualize ditto.umxStandardizeACEplot using DiagrammeR!umx_set_plot_format("DiagrammeR")umx_set_plot_format("graphviz")umx_set_table_format("markdown")umx_set_table_format("latex")umx_set_auto_plot(TRUE)umxEFA
(beta!)loadings() support for umxEFAumx_print, umx_show,
umxCompareplot now works on Windows and Unix as well as OS
X! (h/t @mhunter)umxReRun to umxModify (h/t @hmaes)plot should generate pdfs for Windows and Unix
users nowumxACEcov support for ACE with covariatesumx_make_TwinData added to support simulation
studiesumx_check_parallel now returns timing
informationumx_time now supports “start” and “stop”)umxAPA can format more objects in APA style,
inc just a p-value)summaryAPA can now summarize a data.frame with
mean, sd, and correlationsumxLabel now preserves existing labels in
umxRAM models: labels = is now savedvcov() addedumxRAM can take existing model as input (very
useful)umxRAM reports model fit and comparison (if
old model exists)umxPath now lbounds var@0 for
v.m.umx_residualize can now take multiple
DVsumxMI enhanced to cope with more
situationsumxSummary more resilient with missing CIs
(#ht Nathan Gillespie!)summaryAPA now handles data as well as
lmumx_print resilient to 0-row inputumx_fun_mean_sd)umxCP twin models! + plot(), umxSummary()umxIP twin models! + plot(), umxSummary()umxGxE twin models! + plot(), umxSummary()umxPath limitation where “to” was not being
set.umx_lower2fullconfint(); plot() etc functions
implemented