Package for Environmental Statistics, Including US EPA Guidance

Graphical and statistical analyses of environmental data, with focus on analyzing chemical concentrations and physical parameters, usually in the context of mandated environmental monitoring. Major environmental statistical methods found in the literature and regulatory guidance documents, with extensive help that explains what these methods do, how to use them, and where to find them in the literature. Numerous built-in data sets from regulatory guidance documents and environmental statistics literature. Includes scripts reproducing analyses presented in the book "EnvStats: An R Package for Environmental Statistics" (Millard, 2013, Springer, ISBN 978-1-4614-8455-4, < http://www.springer.com/book/9781461484554>).


News

Changes in Version 2.3.0, 2017.10.09:

o Updated the function gofTestCensored() to peform a goodness-of-fit test 
  using censored data for any continuous distribution that has an associated 
  estimation method for censored data.

o Added the function distChooseCensored() to automatically choose a distribution based on 
  a series of goodness-of-fit tests performed by calling the gofTestCensored() function.

o Added the function distChoose() to automatically choose a distribution based on 
  a series of goodness-of-fit tests performed by calling the gofTest() function.

o For the function egamma():
  --Now allows computation of one-sided lower and one-sided upper 
    confidence intervals when ci.method="profile.likelihood".
  --Added the option ci.method="chisq.adj", which uses the method of 
    Grice and Bain (1980) to compute a confidence interval for the mean.
    This is equivalent to the "adjusted gamma" method of ProUCL. 
  --Added the data object Grice.Bain.80.mat, which is used by egamma() when 
    ci.method="chisq.adj".  (See the help file for Grice.Bain.80.mat.)
  --Fixed the help file for the function egamma() so that it correctly states 
    that when method="bcmle" the scale parameter is computed based on the 
    bias-corrected mle of the shape parameter. 

o For the function gofTest():  added the options test="ad" (Anderson-Darling), 
  test="cvm" (Cramer-von Mises), test="lillie" (Lilliefors), 
  test="ProUCL.ad.gamma" and test="ProUCL.ks.gamma".  
  See the help file for gofTest() for more information.

o Due to issues with numerical integration, rewrote the functions 
  evNormOrdStats() and evNormOrdStatsScalar().  
  The argument 'approximate' has been replaced with the argument 'method', 
  although the argument 'approximate' is still available to use for 
  for backwards compatibility.  See the help file for more details.  
  I am grateful to Gwern Branwen (www.gwern.net) for bringing 
  these issues to my attention. 

o Fixed a typographical error in Equation (4) of the help file for 
  evNormOrdStats() and evNormOrdStatsScalar().

o Fixed minor bug in the function gofTest() that had resulted in an 
  incorrect names attribute for the "data" component of the returned list.

o The following ggproto objects have been exported so that advanced users who are 
  curious may see them.
  --StatNText:             Called by stat_n_text.
  --StatMeanSDText:        Called by stat_mean_sd_text.
  --StatMedianIQRText:     Called by stat_median_iqr_text.
  --StatTestText:          Called by stat_test_text. 

========================

Changes in Version 2.2.1, 2017.01.13:

o Added the function geom_stripchart(), which is a geom that works under the package ggplot2 
  and is the analogue to the EnvStats function stripChart().  Also added the following 
  supporting functions for geom_stripchart():
  --stat_n_text:           beneath each group, displays text indicating the sample size.
  --stat_mean_sd_text:     above each group, displays text indicating the mean and standard deviation.
  --stat_median_iqr_text:  above each group, displays text indicating the median and interquartile range.
  --stat_test_text:        at top of panel, displays text indicating the test results comparing groups. 

o Updated the function eqnpar():
  -- Added the argument "type" to allow the user to specify how the estimated 
     quantile is computed.  This argument is passed to the quantile() function.
  -- Added the option ci.method="interpolate" (now the default).
  -- Changed the behavior when ci.method="exact".
  -- Changed the behavior when ci.method="normal.approx".

o Fixed bug in the function gofTestCensored() for the case when there is only 
  one censoring level but the sample is still considered to be multiply censored 
  because the censoring level is 
  larger than the smallest uncensored observation (left-censored data) or 
  smaller than the largest uncensored observation (right-censored data).

o Updated the help files for enormCensored(), elnormCensored(),  
  elnormAltCensored(), and gofTestCensored():
  -- The default value for prob.method is prob.method="hirsch-stedinger", 
     not "michael-schucany".

========================

Changes in Version 2.1.1, 2016.06.13:

o Updated the function stripChart(): 
  -- For the case when there are two independent groups and 
     plot.diff=TRUE and paired=TRUE:
     ++ Adjusted the default value of diff.lim.
     ++ Added the argument diff.method to indicate what method to use to 
        plot the paired differences.  The default value is diff.method="stack".
  -- For the case when there are two independent groups and 
     plot.diff=TRUE and paired=FALSE:
     ++ The default value for the argument diff.lim (which determines the 
        range on the axis associated with the confidence interval for the 
        difference between groups) is now on the same scale as the range of 
        the original observations.
     ++ The default location of the confidence interval for the difference 
        between groups has changed slightly.
  -- Fixed a bug for the case when ci.and.test='parametric' and the argument 
     test.arg.list contains components named 'paired' and 'var.equal'.
  -- When the argument test.arg.list is supplied with a component named 
     'paired' and the value of that component does not match the value of 
     the argument 'paired', the value of the component is set to the value of 
     the argument 'paired' and a warning is issued.
  -- Fixed a bug for the case when add=TRUE.

o Updated the function summaryStats():
  -- Takes the new argument 'paired' applicable to the case when there are 2 groups 
     that are paired observations. 
     ++ When paired=TRUE and p.value=TRUE, the paired t-test 
        or paired Wilcoxon test is performed.
  -- Fixed bug for the case when there are 2 independent groups, test="parametric", 
     and the argument 'test.arg.list' contains the component paired=TRUE.

o Updated the functions enormCensored() and elnormCensored():
  -- The argument 'method' now accepts the values "ROS" and "rROS" 
     (i.e., you can set method="ROS" or method="rROS").  
     ++ The ROS method denotes the Regression on Order Statistics method.  
        This method is equivalent to the already existing method 
        Quantile-Quantile Regression (method ="qq.reg").
     ++ The rROS method denotes the robust Regression on Order Statistics method.  
        This method is equivalent to the already existing method 
        Imputation Using Quantile-Quantile Regression (method ="impute.w.qq.reg").

o Updated the function elnormAltCensored():
  -- The argument 'method' now accepts the value "rROS" (i.e., you can set 
     method="rROS"), denoting the robust Regression on Order Statistics method.  
     This method is equivalent to the already existing method 
     Imputation Using Quantile-Quantile Regression (method ="impute.w.qq.reg").

========================

Changes in Version 2.1.0, 2016.04.18:

o Added the datasets ACE.13.TCE.df and Helsel.Hirsch.02.Mayfly.df.

o Updated summaryStats():
  -- Fixed bug so it does not return an error when the argument 'object' is a 
     logical or character vector.
  -- Fixed bug so it uses the argument test.arg.list in the case when there 
     are 2 groups, test="parametric", and group.p.value.type="within".
  -- When p.value=TRUE and there are 2 groups, added estimate of difference in 
     location to output:  
     when test="parametric" shows the difference between the means, and  
     when test="nonparametric" shows the median of the difference between 
     a sample from group 1 and a sample from group 2.

o Updated stripChart(): 
  -- Takes new argument 'plot.diff' applicable to the case when there are 2 groups.  
     ++ When plot.diff=TRUE and paired=FALSE, the confidence interval 
        for the difference between the two locations is displayed and the right axis 
        (when vertical=TRUE) or top axis (when vertical=FALSE) is labeled in units of 
        the confidence interval for the difference between the two locations. 
     ++ When plot.diff=TRUE and paired=TRUE, the paired differences are 
        displayed and the right axis (when vertical=TRUE) or 
        top axis (when vertical=FALSE) is labeled in units of the paired differences.  
        In addition, if show.ci=TRUE, the confidence interval based on the paired 
        differences is displayed. 

  -- Takes new argument 'paired' applicable to the case when there are 
     2 groups that are paired observations.  
	++ When paired=TRUE and p.value=TRUE, the paired t-test 
	   or paired Wilcoxon test is performed.

  -- Takes new argument 'paired.lines'.  
     When paired=TRUE and paired.lines=TRUE, lines are drawn between the 
     paired observations.

  -- Takes new argument 'paired.lty'.  
     When paired=TRUE and paired.lines=TRUE, paired.lty is a vector indicating 
     the line types to use to distinguish different paired observations.

  -- Takes new argument 'paired.lwd'.  
     When paired=TRUE and paired.lines=TRUE, paired.lwd is a scalar or vector 
         indicating the width of the lines used to distinguish different paired observations.

  -- Takes new argument 'paired.pch'.  
     When paired=TRUE and paired.lines=TRUE, paired.pch is a vector indicating 
     the plotting symbols to use to distinguish different paired observations.

  -- Takes new argument 'paired.col'.  
     When paired=TRUE and paired.lines=TRUE, paired.col is a vector indicating 
     the color of the lines and/or plotting symbols to use to distinguish 
     different paired observations.  This argument overrides the argument 'col'.

  -- Takes new argument 'diff.col'.
     ++ When plot.diff=TRUE and paired=TRUE, diff.col indicates the color 
        of the displayed paired differences and the color of the confidence 
        interval when show.ci=TRUE.
     ++ When plot.diff=TRUE and paired=FALSE, diff.col indicates the color 
        of the confidence interval for the difference between the two means 
        when show.ci=TRUE.
     ++ This argument overrides the argument 'col'.

  -- Takes new argument 'diff.pch'.
     ++ When plot.diff=TRUE and paired=TRUE, diff.pch indicates the plotting  
        symbol for the displayed paired differences.
     ++ This argument overrides the argument 'pch'.

  -- Takes new arguments 'diff.name', 'diff.name.cex', and 'diff.axis.label'.
     ++ When plot.diff=TRUE, diff.name is a character string indicating the 
        the label for the difference between the two groups.  The default value 
        is Group 2 Name - Group 1 Name where Group 1 Name and Group 2 Name indicate 
        the name of the first and second groups, respectively.
     ++ When plot.diff=TRUE, diff.name.cex is a numeric scalar indicating 
        what value of cex to use to plot the character indicating the estimate of 
        of the difference between groups 1 and 2.
     ++ When plot.diff=TRUE, diff.axis.label is a character string indicating the 
        the label for the axis associated with the difference between the two groups.  
        The default value is "Difference Between Groups" when paired=FALSE, and 
            "Paired Difference" when paired=TRUE.

  -- Takes new arguments 'diff.lim' and 'diff.at'.
     ++ When plot.diff=TRUE, diff.lim is a numeric vector of length 2 indicating the 
            limits to use for the axis associated with the difference between the two groups.
     ++ When plot.diff=TRUE, diff.at is a numeric vector indicating where to place  
            tick marks on the axis associated with the difference between the two groups.

  -- Takes new arguments 'sep.line', 'sep.lty', 'sep.lwd', and 'sep.col'.
     ++ When plot.diff=TRUE, sep.line is a logical scalar indicating whether to 
        draw a line between the second group and the difference between groups 1 and 2.
        The default value is sep.line=TRUE.
     ++ When plot.diff=TRUE, sep.lty is a numeric scalar indicating the line type to 
        use to draw the line between the second group and difference between groups 
        1 and 2.  The default value is sep.lty=2.
     ++ When plot.diff=TRUE, sep.lwd is a numeric scalar indicating the width of the 
        line to draw between the second group and difference between groups 
        1 and 2.  The default value is sep.lwd=cex.
     ++ When plot.diff=TRUE, sep.col is a numeric scalar or character string indicating 
        the color of the line to draw between the second group and difference between 
        groups 1 and 2.  The default value is sep.col="gray".

  -- Takes new arguments 'plot.diff.mar'.
     ++ When plot.diff=TRUE and add = FALSE, plot.diff.mar is a numeric vector of 
        length 4 indicating the value of the mar graphics parameter to use for the plot.  
        The default value is c(5.1, 4.1, 4.1, 4.1), which differs from the usual 
        default value in that the value for side 4 is 4.1 instead of 2.1.  This allows 
        for room to add an axis and axis label on side 4 indicating that that axis 
        represents the difference between the 2 groups.  

o Updated help files and fixed formatting issues in the PDF version.

========================

Changes in Version 2.0.2, 2015.11.17:

o Updated rosnerTest() and help file:
  -- The minimum number of non-missing, finite values in the argument x 
     can now be 3 instead of 10.
  -- An updated version of Rosner's (1983) Table 1 is given in the help file.  
     The table shows the observed Type I error levels for various combinations of 
     sample sizes and values of k. 
  -- The conditions for when a warning is issued when warn=TRUE have been 
     slightly altered.  See the help file for more information.

o Fixed bugs in stripChart.default() so that
  -- The title indicates the correct test when the argument 
     test.arg.list includes the component paired=TRUE.
  -- Does not return an error when the first argument is not a 
     formula and there are missing values and p.value=TRUE.  

o Updated help files and fixed formatting issues in the PDF version.

========================

Changes in Version 2.0.1, 2015.05.24:

o Updated stripChart() by adding the arguments 
  ci.bar.lwd and nsmall.  
  See the help file for details.

o Fixed bug in summaryStats.formula() so that it will correctly 
  deal with a character or logical vector as its first argument.

o Fixed bug in errorBar() so that it does not pass high-level 
  graphics arguments to segments().

o Updated help files and fixed formatting issues in the PDF version.

========================

Changes in Version 2.0.0, 2015.03.09:

o EnvStats contains a modified version of the R function predict.lm().  
  To comply with CRAN policy, added the generic function predict() and the 
  function predict.default(), and also updated the already existing (within 
  EnvStats) predict.lm() function.

o EnvStats contains a modified version of the R function print.htest().  
  To comply with CRAN policy, added the generic function print() and the 
  function print.default().

o Included the function errorBar() in the NAMESPACE.  This function allows 
  the user to plot error bars.  See the help file for more information.

o Updated the function calibrate() to return an object of class "calibrate" 
  which inherits from class "lm".

o Included the function print.gofOutlier() in the NAMESPACE so that the results 
  print correctly for the function rosnerTest() to perform Rosner's test for 
  outliers assuming a normal (Gaussian) distribution.

o Updated summaryStats() so that it will accept a character or logical vector 
  as its first argument.

o Updated stripChart() so that it will accept a numeric matrix as its 
  first argument.

o Fixed a bug in the function elnorm3() for the case when method="lmle".
  (Thanks to Jon Hosking for pointing this out!).

o Fixed formatting problems in the PDF version of the help files.

========================

Changes in Version 1.0.3, 2014.10.20:

o Added the function rosnerTest to perform Rosner's test for outliers in a 
  normal (Gaussian) distribution.

o Modified the function stripChart():
	- The argument ci.offset is now allowed to be either a scalar or 
	  a vector of length equal to the number of groups.
	- A new argument group.names.cex has been added to give explicit 
	  control over the size of the group labels.

o Fixed a bug in gofTestCensored() for the case when the data contain only 
  one censored observation and the censored observation is larger than 
  the smallest non-censored observation for left-censored data or is smaller than 
  the largest non-censored observation for right-censored data. 

o Fixed a bug in anovaPE() that allowed the user to attempt to compute a 
  lack-of-fit and pure error anova table when there are not enough degrees of 
  freedom to do so.

o Fixed a bug in calibrate() to allow only models with enough degrees of 
  freedom to be tested for lack-of-fit via anovaPE.  Also added the arguments 
  test.higher.orders and F.test.

o Modified detectionLimitCalibrate() as follows:

  + now returns not just the detection limit but both the decision limit 
    (on the scale of the signal) and the detection limit (on the scale of 
    the concentration).

  + changed the default value for the argument simultaneous to simultaneous=TRUE.

o Fixed a bug in signTest: values equal to the the argument mu were not 
      being discarded.  (Thanks to Ed Gilroy for pointing this out!).

o Fixed a bug in enparCensored: error message appeared when 
  ci.method="bootstrap".  (Thanks to Dennis Helsel for pointing this 
  out!).

o Fixed a bug in enormCensored when method="mle", ci=TRUE, 
  ci.method="normal.approx", and ci.type="lower" or ci.type="upper":   
  LCL and UCL were switched.  (Thanks to Dennis Helsel for pointing 
  this out!).

o For enormCensored, elnormCensored, elnormAltCensored, egammaCensored, 
  egammaAltCensored, and epoisCensored, one-sided confidence intervals are 
  now available when ci.method="profile.likelihood". 

o For enparCensored, enormCensored, elnormCensored, elnormAltCensored, 
  egammaCensored, egammaAltCensored, and epoisCensored, arguments and results 
  returned when ci.method="bootstrap" have been updated: 

  + The argument use.acc.con has been deprecated.  Bias-corrected 
    and accelerated (BCa) bootstrap results are now always computed 
    where the acceleration constant is estimated rather than allowing 
    the user to set it to 0 (i.e., setting use.acc.con=FALSE in 
    older versions of EnvStats).  Note that 1) code that explicitly 
    set use.acc.con=FALSE will no longer work for these functions, and 
    2) use.acc.con=FALSE was the default so results run previously 
    will not match what is now returned (even if the random seed is 
    set to the same value as was used before).  

  + The argument ci.type is now allowed to be "lower" or "upper".

  + As before, there are two sets of confidence limits that are returned:

      * percentile confidence limits (denoted Pct.LCL and Pct.UCL) and 

      * bias-corrected and accelerated confidence limits (denoted 
        BCa.LCL and BCa.UCL)

  + For enparCensored, a third set of confidence limits are also returned:

      * studentized confidence limits (denoted t.LCL and t.UCL)


o Fixed bugs in oneSamplePermutationTest and twoSamplePermutationTestLocation:
  the default value of the argument seed is now set to seed=NULL.

o Fixed a bug in kendallSeasonalTrendTest: an error message appeared when 
  the argument 'y' was supplied as a matrix and independent.obs=FALSE.

o Change all calls to the EnvStats global variable Distribution.df to 
  EnvStats::Distribution.df.

o Fixed formatting problems in the PDF version of the help files.

========================

Changes in Version 1.0.2, 2013.10.29:

o Changed the names of functions used for minimization in the following functions:
  
  ciBinomN.vec, elnorm3, enormMultiplyCensored.mle, enormSinglyCensored.mle, 
  predIntPois

========================

Changes in Version 1.0.1:

o Fixed bugs in the following functions that had misspelled variable 
  names or variables that had not been defined:

  ci.gamma.chisq.approx, ci.lnorm3.zero.skew, ciBinomN, eqlnorm, 
  plot.gofCensored, ppccNormMultiplyCensoredGofTest, 
  ppccNormSinglyCensoredGofTest, predIntPois, print.boxcox, 
  print.boxcoxCensored

========================

Changes in Version 1.0.0

o The current version of EnvStats is Version 1.0.0.

o EnvStats is built upon Version 2.0 of the S-PLUS module "EnvironmentalStats for S-PLUS".  It includes numerous additions and changes.

o Function and data object names in EnvStats differ from those in EnvironmentalStats for S-PLUS.  All objects that begin with an upper case letter are data objects, and all functions begin with a lower case letter.  So for example, all data objects that began with "epa." in EnvironmentalStats for S-PLUS begin with "EPA." in EnvStats.

o For former EnvironmentalStats for S-PLUS users, here are the name conversions for the functions:

	Old Name						New Name
	--------						--------
	anova.pe						anovaPE
	aov.n							aovN
	aov.power						aovPower
	boxcox.multiply.censored				Embeded in boxcoxCensored
	boxcox.singly.censored					Embeded in boxcoxCensored
	boxcox.transform					boxcoxTransform
	cdf.compare						cdfCompare
	cdf.compare.censored					cdfCompareCensored
	cdfplot							cdfPlot
	chen.t.test						chenTTest
	chisq.gof						Embedded in gofTest
	ci.binom.half.width					ciBinomHalfWidth
	ci.binom.n						ciBinomN
	ci.norm.half.width					ciNormHalfWidth
	ci.norm.n						ciNormN
	ci.npar.conf.level					ciNparConfLevel
	ci.npar.n						ciNparN
	detection.limit.calibrate				detectionLimitCalibrate
	dlnorm.alt						dlnormAlt
	dlnorm.mix						dlnormMix
	dlnorm.mix.alt						dlnormMixAlt
	dlnorm.trunc						dlnormTrunc
	dlnorm.trunc.alt					dlnormTruncAlt
	dnorm.mix						dnormMix
	dnorm.trunc						dnormTrunc
	dzmlnorm.alt						dzmlnormAlt
	ecdplot							ecdfPlot
	ecdfplot.censored					ecdfPlotCensored
	elnorm.alt						elnormAlt
	elnorm.alt.multiply.censored				Embeded in elnormAltCensored
	elnorm.alt.singly.censored				Embeded in elnormAltCensored
	elnorm.multiply.censored				Embeded in elnormCensored
	elnorm.singly.censored					Embeded in elnormCensored
	enorm.multiply.censored					Embeded in enormCensored
	enorm.singly.censored					Embeded in enormCensored
	enpar.censored						enparCensored
	epdfplot						epdfPlot
	epois.multiply.censored					Embeded in epoisCensored
	epois.singly.censored					Embeded in epoisCensored
	error.bar						errorBar
	ev.norm.ord.stats					evNormOrdStats
	ev.norm.ord.stats.scalar				evNormOrdStatsScalar
	ezmlnorm.alt						ezmlnormAlt
	full.summary						summaryFull
	geo.mean						geoMean
	geo.sd							geoSD
	pdfplot							pdfPlot
	inverse.predict.calibrate				inversePredictCalibrate
	kendall.trend.test					kendallTrendTest
	ks.gof							Embedded in gofTest
	l.moment						lMoment
	linear.trend.test.n					linearTrendTestN
	linear.trend.test.power					linearTrendTestPower
	linear.trend.test.scaled.mds				linearTrendTestScaledMds
	log.choose.multinomial					logChooseMultinomial
	one.sample.permutation.test				oneSamplePermutationTest
	pdfplot							pdfPlot
	plnorm.alt						plnormAlt
	plnorm.mix						plnormMix
	plnorm.mix.alt						plnormMixAlt
	plnorm.trunc						plnormTrunc
	plnorm.trunc.alt					plnormTruncAlt
	plot.aov.design						plotAovDesign
	plot.ci.binom.design					plotCiBinomDesign
	plot.ci.norm.design					plotCiNormDesign
	plot.ci.npar.design					plotCiNparDesign
	plot.linear.trend.test.design				plotLinearTrendTestDesign
	plot.permutation.test					plot.permutationTest
	plot.pred.int.lnorm.alt.simultaneous.test.power.curve	plotPredIntLnormAltSimultaneousTestPowerCurve
	plot.pred.int.lnorm.alt.test.power.curve		plotPredIntLnormAltTestPowerCurve
	plot.pred.int.norm.design				plotPredIntNormDesign
	plot.pred.int.norm.simultaneous.test.power.curve	plotPredIntNormSimultaneousTestPowerCurve
	plot.pred.int.norm.test.power.curve			plotPredIntNormTestPowerCurve
	plot.pred.int.npar.design				plotPredIntNparDesign
	plot.pred.int.npar.simultaneous.design			plotPredIntNparSimultaneousDesign
	plot.pred.int.npar.simultaneous.test.power.curve	plotPredIntNparSimultaneousTestPowerCurve
	plot.prop.test.design					plotPropTestDesign
	plot.t.test.design					plotTTestDesign
	plot.t.test.lnorm.alt.design				plotTTestLnormAltDesign
	plot.tol.int.norm.design				plotTolIntNormDesign
	plot.tol.int.npar.design				plotTolIntNparDesign
	pnorm.mix						pnormMix
	pnorm.trunc						pnormTrunc
	ppcc.evd.gof						Embedded in gofTest
	ppcc.norm.gof						Embedded in gofTest
	ppcc.norm.group.gof					Embedded in gofGroupTest
	ppcc.norm.multiply.censored.gof				Embedded in gofCensoredTest
	ppcc.norm.singly.censored.gof				Embedded in gofCensoredTest
	ppoints.censored					ppointCensored
	pred.int.lnorm						predIntLnorm
	pred.int.lnorm.alt.simultaneous.test.power		predIntLnormAltSimultaneousTestPower
	pred.int.lnorm.alt.test.power				predIntLnormAltTestPower
	pred.int.lnorm.simultaneous				predIntLnormSimultaneous
	pred.int.norm						predIntNorm
	pred.int.norm.half.width				predIntNormHalfWidth
	pred.int.norm.K						predIntNormK
	pred.int.norm.n						predIntNormN
	pred.int.norm.simultaneous				predIntNormSimultaneous
	pred.int.norm.simultaneous.K				predIntNormSimultaneousK
	pred.int.norm.simultaneous.test.power			predIntNormSimultaneousTestPower
	pred.int.norm.test.power				predIntNormTestPower
	pred.int.npar						predIntNpar
	pred.int.npar.conf.level				predIntNparConfLevel
	pred.int.npar.n						predIntNparN
	pred.int.npar.simultaneous				predIntNparSimultaneous
	pred.int.npar.simultaneous.conf.level			predIntNparSimultaneousConfLevel
	pred.int.npar.simultaneous.n				predIntNparSimultaneousN
	pred.int.npar.simultaneous.test.power			predIntNparSimultaneousTestPower
	pred.int.pois						predIntPois
	prop.test.mdd						propTestMdd
	prop.test.n						propTestN
	prop.test.power						propTestPower
	pw.moment						pwMoment
	pzmlnorm.alt						pzmlnormAlt
	qlnorm.alt						qlnormAlt
	qlnorm.mix						qlnormMix
	qlnorm.mix.alt						qlnormMixAlt
	qlnorm.trunc						qlnormTrunc
	qlnorm.trunc.alt					qlnormTruncAlt
	qnorm.mix						qnormMix
	qnorm.trunc						qnormTrunc
	qqline							qqLine
	qqplot							qqPlot
	qqplot.gestalt						qqPlotGestalt
	qqplot.censored						qqPlotCensored
	quantile.test						quantileTest
	qzmlnorm.alt						qzmlnormAlt
	rlnorm.alt						rlnormAlt
	rlnorm.mix						rlnormMix
	rlnorm.mix.alt						rlnormMixAlt
	rlnorm.trunc						rlnormTrunc
	rlnorm.trunc.alt					rlnormTruncAlt
	rnorm.mix						rnormMix
	rnorm.trunc						rnormTrunc
	rzmlnorm.alt						rzmlnormAlt
	seasonal.kendall.trend.test				kendallSeasonalTrendTest
	serial.correlation.test					serialCorrelationTest
	sf.gof							Embedded in gofTest
	sf.group.gof						Embedded in gofGroupTest
	sf.multiply.censored.gof				Embedded in gofCensoredTest
	sf.singly.censored.gof					Embedded in gofCensoredTest
	sign.test						signTest
	simulate.mv.matrix					simulateMvMatrix
	simulate.vector						simulateVector
	skew.gof						Embedded in gofTest
	sw.gof							Embedded in gofTest
	sw.group.gof						Embedded in gofGroupTest
	sw.singly.censored.gof					Embedded in gofCensoredTest
	t.test.n						tTestN
	t.test.power						tTestPower
	t.test.scaled.mdd					tTestScaledMdd
	t.test.lnorm.alt.n					tTestLnormAltN
	t.test.lnorm.alt.power					tTestLnormAltPower
	t.test.lnorm.alt.ratio.of.means				tTestLnormAltRatioOfMeans
	tol.int.lnorm						tolIntLnorm
	tol.int.norm						tolIntNorm
	tol.int.norm.half.width					tolIntNormHalfWidth
	tol.int.norm.K						tolIntNormK
	tol.int.norm.n						tolIntNormN
	tol.int.npar						tolIntNpar
	tol.int.npar.conf.level					tolIntNparConfLevel
	tol.int.npar.coverage					tolIntNparCoverage
	tol.int.npar.n						tolIntNparN
	tol.int.pois						tolIntPois
	two.sample.linear.rank.test				twoSampleLinearRankTest
	two.sample.linear.rank.test.censored			twoSampleLinearRankTestCensored
	two.sample.permutation.test.location			twoSamplePermutationTestLocation
	two.sample.permutation.test.proportion			twoSamplePermuationTestProportion
	var.group.test						varGroupTest
	var.test						varTest
	z.test.gevd.shape					zTestGevdShape

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.

install.packages("EnvStats")

2.3.0 by Steven P. Millard, 4 months ago


http://www.probstatinfo.com


Browse source code at https://github.com/cran/EnvStats


Authors: Steven P. Millard <[email protected]>


Documentation:   PDF Manual  


Task views: Analysis of Ecological and Environmental Data, Probability Distributions


GPL (>= 3) license


Imports MASS, ggplot2, nortest

Suggests lattice, qcc, sp, boot


Imported by AnnuityRIR, WRTDStidal, elementR, eva, simPop.

Suggested by fitteR, httk.


See at CRAN