Log message:
(math/R-car) Updated 3.1.2 to 3.1.3

Changes to Version 3.1-3

  o Clarify docs to explain that ... argument is not passed to .vcov
  when it's a function in calls to Anova(), confidence Elliipse(),
  Confint(), deltaMethod(), linearHypothesis(), and S() (suggestion of
  Joe Ritter).

  o make dataEllipse() generic and add default and formula methods
  (with help of Michael Friendly).

  o linearHypothesis() no longer issues a warning when a hypothesis
  matrix is supplied for coefficients whose names include arithmetic
  operators (following report by Robert Baer about unnecessary
  warnings printed by Anova(), which calls linearHypothesis()).

  o deltaMethod() now allows models with aliased coefficients when
  argument singular.ok=TRUE.

  o Anova() method for "coxph" objects no longer computes incorrect LR
  tests for models using the tt argument; in this case, Wald tests are
  substituted (following bug report by Ken Beath).

  o crPlot(), crPlots(), crPlot3d(), and avPlot3d() now invisibly
  return coordinates rather than NULL, as avPlot() and avPlots()
  already did (following query by Jeff Reichman).

  o poTest() supports "polr" models with weights (suggestion of Ken
  Beath).

  o Anova() now supports "clm" and "clmm" objects created by \ 
functions
  in the ordinal package (after report by Karl Ove Hufthammer that the
  default method doesn't work).

  o Anova.default() does more checking for potential problems (after
  report by Karl Ove Hufthammer).

  o Anova.lme() correctly handles models with contrasts different from
  the default contr.treatment (fixing a bug reported by Ben Bolker).

  o Small changes to docs.

Log message:
(math/R-car) Update 3.1.1 to 3.1.2

Changes to Version 3.1-2

  o Fixed bugs in interface for residualPlots() and residualPlot()

  o Fixed hccm() to return a correct error message when regressors are
   ill-conditioned.

  o Use t-distribution for score tests in boxTidwell() and add omnibus
   score test (suggestions of Peter Dalgaard).

  o residualPlots() is no longer confused by a one-column matrix
    regressor, produced, e.g., by scale() (issue reported by Tom
    Wenseleers).

  o Make the printed formating of hypotheses in linearHypothesis()
  more tolerant of coefficients with certain non-standard names (after
  issue reported by Matthieu Stigler).

  o Tweak linearHypothesis() for multinom models and add example.

  o Update scatter3d() and crPlot3d() to use rgl::*3d() functions;
    requires rgl (>= 0.111.3) (patch courtesy of Duncan Murdoch).

Log message:
(math/R-car) Updated 3.0.13 to 3.1.1

Changes to Version 3.1-1

  o Fixed bug in hccm(): when some cases are fit exactly all versions
    of the hccm estimate are singular and inconsistent.  A error message
    is now returned.

  o Fixed bug in Anova.survreg() for Wald tests (reported by Megan
    Taylor Jones), which failed with a spurious "missing residual df"
    error.

  o Made Anova.lm() more robust when there are aliased coefficients
    (following report of cryptic output by Taiwo Fagbohungbe).

  o Added reg.function, reg.function.col, and mouseMode arguments to
    scatter3d(), and mouseMode to crPlot3d().

  o Added fill and associated arguments to influencePlot() (initial
    implementation of Michael Friendly).

  o Added confidenceEllipses() function to create matrix of all
    pairwise confidence ellipses for a regression model (suggestion of
    Michael Friendly).

  o Moved the pointLabel() function from the maptools package in
    anticipation of the retirement of that package, and with the
    permission of the function author, Tom Short.

Changes to Version 3.1-0

  o New 3D added-variable and component+residual plots, avPlot3d() and
    crPlot3d().

  o More flexible recode()/Recode().

  o Added Anova.svycoxph() to prevent selection of test.statistic="LR".

  o New take on vif.lm().

Log message:
(math/R-car) Updated 3.0.11 to 3.0.13

Changes to Version 3.0-13

  o Handle singularities better in Anova.mlm() (suggestion of Marius Barth).

  o Fixed error reporting in logit() (bug reported by Collin Erickson).

  o Added pt.cex and cex sub-arguments to legend argument of
  scatterplotMatrix() (implicit suggestion of Duncan Mackay).

  o New vif.lm() method with options for handling linear models with interactions.

  o Fixed bug in cooks.distance.influence.lme() (in effect, incorrect
  division by the error variance twice rather than once), and a sign
  error in dfbeta.influence.lme() and dfbetas.influence.lme()
  (following report by Ben Bolker).

  o Added an error.df argument to the default methods of
  linearHypothesis() and Anova() for computing F-tests; if the
  argument is unspecified, then df.residual() is applied to the model,
  and if df.residual() returns NULL or NA, then an asymptotic
  chi-square test is substituted for the F-test (in response to a
  report by Adrian Olszewski that Anova() fails for gls models when an
  F-test is requested).

Changes to Version 3.0-12

  o The 'rio` package is now suggested, not required.  To use the
  Import() and Export() functions users may need to install rio using
  install.packages("rio")

  o hccm() now reports an interpretable error, except for type = "hc0"
  or "hc1", when there are cases with hatvalues = 1 (following problem
  reported by Peng Ding).

  o Fix bug in rownames in dfbetasPlots.lm() (reported by Shu Fai
  Cheun).

  o Package 'rio' was moved from Imports to Suggests. Both Import()
  and Export() now check for the availability of 'rio' first and
  report an error otherwise.

Log message:
math: Replace RMD160 checksums with BLAKE2s checksums

All checksums have been double-checked against existing RMD160 and
SHA512 hashes

Log message:
math: Remove SHA1 hashes for distfiles

Log message:
(math/R-car) Updated 3.0.10 to 3.0.11

Changes to Version 3.0-11

  o boxCox() function now allows the use of any of the usual graphics
    parameters in the plot() function.  If particuar boxCox(model,
    main="my title") will add a title to the plot, and boxCox(model,
    main="") will suppress it.

  o car::Boot(object, method="residual") will work for
    method="residual" only if the car package has been previously
    attached via either library(car) or require(car).

  o Added error checking with vcov. argument.  Changed vcov.boot to
    print a warning of bootstrap replicates that returned NA

  o Introduced brief.tbl(), which simply calls print(), to cope with
    changes to tibbles.

  o qqPlot() fills (shades) confidence envelopes by default, and
    smoothers for scatterplots fill variance (spread) envelopes by
    default (suggestions of Michael Friendly).

  o Fixed problem in infIndexPlot.influence.lme() (reported by Francis
    L. Huang).

  o New "polr" and "svyolr" methods for vif() (following \ 
report by
    Abra Jeffers).

  o Make linearHypothesis() and Anova() work with "svyolr" objects via
    their default methods.

  o Regularize handling of vcov. argument in Anova() and
    linearHypothesis().

  o vcov. argument now works with Anova() for models produced by
    lme4::lmer() and glmer(), (fixing a bug reported by Amy MacDougall).

  o New linearHypothesis.lmList() method.

  o New "lm" method for symbox().

  o New cex and pt.wts arguments for avPlot() "lm" and "glm" \ 
methods,
    and for mcPlot.lm().

  o Fix handling of imatrix argument to Anova.mlm() (suggestion of
    Benedikt Langenberg).

  o Remove influence.merMod() and related methods in favor of versions
    of these methods in the lme4 package.

  o Rewrite "embedding" vignette.

  o Small fixes and improvements.

Log message:
Update to 3.0.10

Upstream changes:
Changes to Version 3.0-10

  o Fixed as error when using parameter name "(Intercept)" with \ 
deltaMethod().

  o Several changes to Boot() to correct bugs, support the plinear algorithm, to \ 
correct use of weights for lm, and allow use of weights for nls.

  o Fix quantregLine() to work with development version 5.69 of quantreg package \ 
(following heads-up from Roger Koenker).