R/linreg.h.R
In jamovi/Rjamovi: The 'jamovi' Analyses
Defines functions
linReg
# This file is automatically generated, you probably don't want to edit this
linRegOptions <- if (requireNamespace("jmvcore", quietly=TRUE)) R6::R6Class(
"linRegOptions",
inherit = jmvcore::Options,
public = list(
initialize = function(
dep = NULL,
covs = NULL,
factors = NULL,
weights = NULL,
blocks = list(
list()),
refLevels = NULL,
intercept = "refLevel",
r = TRUE,
r2 = TRUE,
r2Adj = FALSE,
aic = FALSE,
bic = FALSE,
rmse = FALSE,
modelTest = FALSE,
anova = FALSE,
ci = FALSE,
ciWidth = 95,
stdEst = FALSE,
ciStdEst = FALSE,
ciWidthStdEst = 95,
norm = FALSE,
qqPlot = FALSE,
resPlots = FALSE,
durbin = FALSE,
collin = FALSE,
cooks = FALSE,
mahal = FALSE,
mahalp = "0.001",
emMeans = list(
list()),
ciEmm = TRUE,
ciWidthEmm = 95,
emmPlots = TRUE,
emmTables = FALSE,
emmWeights = TRUE, ...) {
super$initialize(
package="jmv",
name="linReg",
requiresData=TRUE,
...)
private$..dep <- jmvcore::OptionVariable$new(
"dep",
dep,
suggested=list(
"continuous"),
permitted=list(
"numeric"))
private$..covs <- jmvcore::OptionVariables$new(
"covs",
covs,
suggested=list(
"continuous"),
permitted=list(
"numeric"),
default=NULL)
private$..factors <- jmvcore::OptionVariables$new(
"factors",
factors,
rejectUnusedLevels=TRUE,
suggested=list(
"nominal"),
permitted=list(
"factor"),
default=NULL)
private$..weights <- jmvcore::OptionVariable$new(
"weights",
weights,
suggested=list(
"continuous"),
permitted=list(
"numeric"),
default=NULL)
private$..blocks <- jmvcore::OptionArray$new(
"blocks",
blocks,
default=list(
list()),
template=jmvcore::OptionTerms$new(
"blocks",
NULL))
private$..refLevels <- jmvcore::OptionArray$new(
"refLevels",
refLevels,
items="(factors)",
default=NULL,
template=jmvcore::OptionGroup$new(
"refLevels",
NULL,
elements=list(
jmvcore::OptionVariable$new(
"var",
NULL,
content="$key"),
jmvcore::OptionLevel$new(
"ref",
NULL))))
private$..intercept <- jmvcore::OptionList$new(
"intercept",
intercept,
options=list(
"refLevel",
"grandMean"),
default="refLevel")
private$..r <- jmvcore::OptionBool$new(
"r",
r,
default=TRUE)
private$..r2 <- jmvcore::OptionBool$new(
"r2",
r2,
default=TRUE)
private$..r2Adj <- jmvcore::OptionBool$new(
"r2Adj",
r2Adj,
default=FALSE)
private$..aic <- jmvcore::OptionBool$new(
"aic",
aic,
default=FALSE)
private$..bic <- jmvcore::OptionBool$new(
"bic",
bic,
default=FALSE)
private$..rmse <- jmvcore::OptionBool$new(
"rmse",
rmse,
default=FALSE)
private$..modelTest <- jmvcore::OptionBool$new(
"modelTest",
modelTest,
default=FALSE)
private$..anova <- jmvcore::OptionBool$new(
"anova",
anova,
default=FALSE)
private$..ci <- jmvcore::OptionBool$new(
"ci",
ci,
default=FALSE)
private$..ciWidth <- jmvcore::OptionNumber$new(
"ciWidth",
ciWidth,
min=50,
max=99.9,
default=95)
private$..stdEst <- jmvcore::OptionBool$new(
"stdEst",
stdEst,
default=FALSE)
private$..ciStdEst <- jmvcore::OptionBool$new(
"ciStdEst",
ciStdEst,
default=FALSE)
private$..ciWidthStdEst <- jmvcore::OptionNumber$new(
"ciWidthStdEst",
ciWidthStdEst,
min=50,
max=99.9,
default=95)
private$..norm <- jmvcore::OptionBool$new(
"norm",
norm,
default=FALSE)
private$..qqPlot <- jmvcore::OptionBool$new(
"qqPlot",
qqPlot,
default=FALSE)
private$..resPlots <- jmvcore::OptionBool$new(
"resPlots",
resPlots,
default=FALSE)
private$..durbin <- jmvcore::OptionBool$new(
"durbin",
durbin,
default=FALSE)
private$..collin <- jmvcore::OptionBool$new(
"collin",
collin,
default=FALSE)
private$..cooks <- jmvcore::OptionBool$new(
"cooks",
cooks,
default=FALSE)
private$..mahal <- jmvcore::OptionBool$new(
"mahal",
mahal,
default=FALSE)
private$..mahalp <- jmvcore::OptionList$new(
"mahalp",
mahalp,
options=list(
"0.05",
"0.01",
"0.001"),
default="0.001")
private$..emMeans <- jmvcore::OptionArray$new(
"emMeans",
emMeans,
default=list(
list()),
template=jmvcore::OptionVariables$new(
"emMeans",
NULL))
private$..ciEmm <- jmvcore::OptionBool$new(
"ciEmm",
ciEmm,
default=TRUE)
private$..ciWidthEmm <- jmvcore::OptionNumber$new(
"ciWidthEmm",
ciWidthEmm,
min=50,
max=99.9,
default=95)
private$..emmPlots <- jmvcore::OptionBool$new(
"emmPlots",
emmPlots,
default=TRUE)
private$..emmTables <- jmvcore::OptionBool$new(
"emmTables",
emmTables,
default=FALSE)
private$..emmWeights <- jmvcore::OptionBool$new(
"emmWeights",
emmWeights,
default=TRUE)
private$..predictOV <- jmvcore::OptionOutput$new(
"predictOV")
private$..residsOV <- jmvcore::OptionOutput$new(
"residsOV")
private$..cooksOV <- jmvcore::OptionOutput$new(
"cooksOV")
private$..mahalOV <- jmvcore::OptionOutput$new(
"mahalOV")
self$.addOption(private$..dep)
self$.addOption(private$..covs)
self$.addOption(private$..factors)
self$.addOption(private$..weights)
self$.addOption(private$..blocks)
self$.addOption(private$..refLevels)
self$.addOption(private$..intercept)
self$.addOption(private$..r)
self$.addOption(private$..r2)
self$.addOption(private$..r2Adj)
self$.addOption(private$..aic)
self$.addOption(private$..bic)
self$.addOption(private$..rmse)
self$.addOption(private$..modelTest)
self$.addOption(private$..anova)
self$.addOption(private$..ci)
self$.addOption(private$..ciWidth)
self$.addOption(private$..stdEst)
self$.addOption(private$..ciStdEst)
self$.addOption(private$..ciWidthStdEst)
self$.addOption(private$..norm)
self$.addOption(private$..qqPlot)
self$.addOption(private$..resPlots)
self$.addOption(private$..durbin)
self$.addOption(private$..collin)
self$.addOption(private$..cooks)
self$.addOption(private$..mahal)
self$.addOption(private$..mahalp)
self$.addOption(private$..emMeans)
self$.addOption(private$..ciEmm)
self$.addOption(private$..ciWidthEmm)
self$.addOption(private$..emmPlots)
self$.addOption(private$..emmTables)
self$.addOption(private$..emmWeights)
self$.addOption(private$..predictOV)
self$.addOption(private$..residsOV)
self$.addOption(private$..cooksOV)
self$.addOption(private$..mahalOV)
}),
active = list(
dep = function() private$..dep$value,
covs = function() private$..covs$value,
factors = function() private$..factors$value,
weights = function() private$..weights$value,
blocks = function() private$..blocks$value,
refLevels = function() private$..refLevels$value,
intercept = function() private$..intercept$value,
r = function() private$..r$value,
r2 = function() private$..r2$value,
r2Adj = function() private$..r2Adj$value,
aic = function() private$..aic$value,
bic = function() private$..bic$value,
rmse = function() private$..rmse$value,
modelTest = function() private$..modelTest$value,
anova = function() private$..anova$value,
ci = function() private$..ci$value,
ciWidth = function() private$..ciWidth$value,
stdEst = function() private$..stdEst$value,
ciStdEst = function() private$..ciStdEst$value,
ciWidthStdEst = function() private$..ciWidthStdEst$value,
norm = function() private$..norm$value,
qqPlot = function() private$..qqPlot$value,
resPlots = function() private$..resPlots$value,
durbin = function() private$..durbin$value,
collin = function() private$..collin$value,
cooks = function() private$..cooks$value,
mahal = function() private$..mahal$value,
mahalp = function() private$..mahalp$value,
emMeans = function() private$..emMeans$value,
ciEmm = function() private$..ciEmm$value,
ciWidthEmm = function() private$..ciWidthEmm$value,
emmPlots = function() private$..emmPlots$value,
emmTables = function() private$..emmTables$value,
emmWeights = function() private$..emmWeights$value,
predictOV = function() private$..predictOV$value,
residsOV = function() private$..residsOV$value,
cooksOV = function() private$..cooksOV$value,
mahalOV = function() private$..mahalOV$value),
private = list(
..dep = NA,
..covs = NA,
..factors = NA,
..weights = NA,
..blocks = NA,
..refLevels = NA,
..intercept = NA,
..r = NA,
..r2 = NA,
..r2Adj = NA,
..aic = NA,
..bic = NA,
..rmse = NA,
..modelTest = NA,
..anova = NA,
..ci = NA,
..ciWidth = NA,
..stdEst = NA,
..ciStdEst = NA,
..ciWidthStdEst = NA,
..norm = NA,
..qqPlot = NA,
..resPlots = NA,
..durbin = NA,
..collin = NA,
..cooks = NA,
..mahal = NA,
..mahalp = NA,
..emMeans = NA,
..ciEmm = NA,
..ciWidthEmm = NA,
..emmPlots = NA,
..emmTables = NA,
..emmWeights = NA,
..predictOV = NA,
..residsOV = NA,
..cooksOV = NA,
..mahalOV = NA)
)
linRegResults <- if (requireNamespace("jmvcore", quietly=TRUE)) R6::R6Class(
"linRegResults",
inherit = jmvcore::Group,
active = list(
modelFit = function() private$.items[["modelFit"]],
modelComp = function() private$.items[["modelComp"]],
models = function() private$.items[["models"]],
predictOV = function() private$.items[["predictOV"]],
residsOV = function() private$.items[["residsOV"]],
cooksOV = function() private$.items[["cooksOV"]],
mahalOV = function() private$.items[["mahalOV"]]),
private = list(),
public=list(
initialize=function(options) {
super$initialize(
options=options,
name="",
title="Linear Regression")
self$add(jmvcore::Table$new(
options=options,
name="modelFit",
title="Model Fit Measures",
clearWith=list(
"dep",
"blocks",
"weights"),
visible="(r || r2 || r2Adj || aic || bic || rmse || modelTest)",
columns=list(
list(
`name`="model",
`title`="Model",
`type`="text"),
list(
`name`="r",
`title`="R",
`type`="number",
`visible`="(r)"),
list(
`name`="r2",
`title`="R\u00B2",
`type`="number",
`visible`="(r2)"),
list(
`name`="r2Adj",
`title`="Adjusted R\u00B2",
`type`="number",
`visible`="(r2Adj)"),
list(
`name`="aic",
`title`="AIC",
`type`="number",
`visible`="(aic)"),
list(
`name`="bic",
`title`="BIC",
`type`="number",
`visible`="(bic)"),
list(
`name`="rmse",
`title`="RMSE",
`type`="number",
`visible`="(rmse)"),
list(
`name`="f",
`title`="F",
`type`="number",
`superTitle`="Overall Model Test",
`visible`="(modelTest)"),
list(
`name`="df1",
`title`="df1",
`type`="integer",
`superTitle`="Overall Model Test",
`visible`="(modelTest)"),
list(
`name`="df2",
`title`="df2",
`type`="integer",
`superTitle`="Overall Model Test",
`visible`="(modelTest)"),
list(
`name`="p",
`title`="p",
`type`="number",
`format`="zto,pvalue",
`superTitle`="Overall Model Test",
`visible`="(modelTest)"))))
self$add(jmvcore::Table$new(
options=options,
name="modelComp",
title="Model Comparisons",
clearWith=list(
"dep",
"blocks",
"weights"),
columns=list(
list(
`name`="model1",
`title`="Model",
`content`=".",
`type`="text",
`superTitle`="Comparison"),
list(
`name`="sep",
`title`="",
`content`="-",
`type`="text",
`format`="narrow",
`superTitle`="Comparison"),
list(
`name`="model2",
`title`="Model",
`content`=".",
`type`="text",
`superTitle`="Comparison"),
list(
`name`="r2",
`title`="\u0394R\u00B2",
`type`="number"),
list(
`name`="f",
`title`="F",
`type`="number"),
list(
`name`="df1",
`title`="df1",
`type`="integer"),
list(
`name`="df2",
`title`="df2",
`type`="integer"),
list(
`name`="p",
`title`="p",
`type`="number",
`format`="zto,pvalue"))))
self$add(jmvcore::Array$new(
options=options,
name="models",
title="Model Specific Results",
layout="listSelect",
hideHeadingOnlyChild=TRUE,
clearWith=list(
"blocks"),
template=R6::R6Class(
inherit = jmvcore::Group,
active = list(
anova = function() private$.items[["anova"]],
coef = function() private$.items[["coef"]],
dataSummary = function() private$.items[["dataSummary"]],
assump = function() private$.items[["assump"]],
emm = function() private$.items[["emm"]]),
private = list(),
public=list(
initialize=function(options) {
super$initialize(
options=options,
name="undefined",
title="")
self$add(jmvcore::Table$new(
options=options,
name="anova",
title="Omnibus ANOVA Test",
visible="(anova)",
refs="car",
clearWith=list(
"dep",
"blocks",
"weights"),
columns=list(
list(
`name`="term",
`title`="",
`type`="text"),
list(
`name`="ss",
`title`="Sum of Squares",
`type`="number"),
list(
`name`="df",
`title`="df",
`type`="integer"),
list(
`name`="ms",
`title`="Mean Square",
`type`="number"),
list(
`name`="F",
`title`="F",
`type`="number"),
list(
`name`="p",
`title`="p",
`type`="number",
`format`="zto,pvalue"))))
self$add(jmvcore::Table$new(
options=options,
name="coef",
title="`Model Coefficients - ${ dep }`",
clearWith=list(
"dep",
"blocks",
"refLevels",
"ciWidth",
"ciWidthStdEst",
"intercept",
"weights"),
columns=list(
list(
`name`="term",
`title`="Predictor",
`type`="text"),
list(
`name`="est",
`title`="Estimate",
`type`="number"),
list(
`name`="se",
`title`="SE",
`type`="number"),
list(
`name`="lower",
`title`="Lower",
`type`="number",
`visible`="(ci)"),
list(
`name`="upper",
`title`="Upper",
`type`="number",
`visible`="(ci)"),
list(
`name`="t",
`title`="t",
`type`="number"),
list(
`name`="p",
`title`="p",
`type`="number",
`format`="zto,pvalue"),
list(
`name`="stdEst",
`title`="Stand. Estimate",
`type`="number",
`visible`="(stdEst)"),
list(
`name`="stdEstLower",
`title`="Lower",
`type`="number",
`visible`="(ciStdEst && stdEst)"),
list(
`name`="stdEstUpper",
`title`="Upper",
`type`="number",
`visible`="(ciStdEst && stdEst)"))))
self$add(R6::R6Class(
inherit = jmvcore::Group,
active = list(
cooks = function() private$.items[["cooks"]],
mahal = function() private$.items[["mahal"]]),
private = list(),
public=list(
initialize=function(options) {
super$initialize(
options=options,
name="dataSummary",
title="Data Summary")
self$add(jmvcore::Table$new(
options=options,
name="cooks",
title="Cook's Distance",
rows=1,
visible="(cooks)",
clearWith=list(
"dep",
"blocks",
"weights"),
columns=list(
list(
`name`="mean",
`title`="Mean",
`type`="number"),
list(
`name`="median",
`title`="Median",
`type`="number"),
list(
`name`="sd",
`title`="SD",
`type`="number"),
list(
`name`="min",
`title`="Min",
`type`="number",
`superTitle`="Range"),
list(
`name`="max",
`title`="Max",
`type`="number",
`superTitle`="Range"))))
self$add(jmvcore::Table$new(
options=options,
name="mahal",
title="Mahalanobis Distance",
rows=1,
visible="(mahal)",
clearWith=list(
"dep",
"blocks",
"weights"),
columns=list(
list(
`name`="mean",
`title`="Mean",
`type`="number"),
list(
`name`="median",
`title`="Median",
`type`="number"),
list(
`name`="sd",
`title`="SD",
`type`="number"),
list(
`name`="min",
`title`="Min",
`type`="number",
`superTitle`="Range"),
list(
`name`="max",
`title`="Max",
`type`="number",
`superTitle`="Range"),
list(
`name`="excRow",
`title`="Participants (rows) below p-threshold",
`type`="text"))))}))$new(options=options))
self$add(R6::R6Class(
inherit = jmvcore::Group,
active = list(
durbin = function() private$.items[["durbin"]],
collin = function() private$.items[["collin"]],
norm = function() private$.items[["norm"]],
qqPlot = function() private$.items[["qqPlot"]],
resPlots = function() private$.items[["resPlots"]]),
private = list(),
public=list(
initialize=function(options) {
super$initialize(
options=options,
name="assump",
title="Assumption Checks")
self$add(jmvcore::Table$new(
options=options,
name="durbin",
title="Durbin\u2013Watson Test for Autocorrelation",
rows=1,
visible="(durbin)",
refs="car",
clearWith=list(
"dep",
"blocks",
"weights"),
columns=list(
list(
`name`="autoCor",
`title`="Autocorrelation",
`type`="number"),
list(
`name`="dw",
`title`="DW Statistic",
`type`="number"),
list(
`name`="p",
`title`="p",
`type`="number",
`format`="zto,pvalue"))))
self$add(jmvcore::Table$new(
options=options,
name="collin",
title="Collinearity Statistics",
visible="(collin)",
refs="car",
clearWith=list(
"dep",
"blocks",
"weights"),
columns=list(
list(
`name`="term",
`title`="",
`type`="text"),
list(
`name`="vif",
`title`="VIF",
`type`="number"),
list(
`name`="tol",
`title`="Tolerance",
`type`="number"))))
self$add(jmvcore::Table$new(
options=options,
name="norm",
title="Normality Test (Shapiro-Wilk)",
visible="(norm)",
rows=1,
clearWith=list(
"dep",
"blocks",
"weights"),
columns=list(
list(
`name`="t[sw]",
`title`="",
`type`="text",
`content`="Shapiro-Wilk",
`visible`=FALSE),
list(
`name`="s[sw]",
`title`="Statistic"),
list(
`name`="p[sw]",
`title`="p",
`format`="zto,pvalue"))))
self$add(jmvcore::Image$new(
options=options,
name="qqPlot",
title="Q-Q Plot",
width=450,
height=400,
renderFun=".qqPlot",
visible="(qqPlot)",
requiresData=TRUE,
clearWith=list(
"dep",
"blocks",
"weights")))
self$add(jmvcore::Array$new(
options=options,
name="resPlots",
title="Residuals Plots",
visible="(resPlots)",
template=jmvcore::Image$new(
options=options,
renderFun=".resPlot",
requiresData=TRUE,
clearWith=list(
"dep",
"blocks",
"weights"))))}))$new(options=options))
self$add(jmvcore::Array$new(
options=options,
name="emm",
title="Estimated Marginal Means",
refs="emmeans",
clearWith=list(
"dep",
"blocks",
"weights",
"emMeans"),
template=R6::R6Class(
inherit = jmvcore::Group,
active = list(
emmPlot = function() private$.items[["emmPlot"]],
emmTable = function() private$.items[["emmTable"]]),
private = list(),
public=list(
initialize=function(options) {
super$initialize(
options=options,
name="undefined",
title="")
self$add(jmvcore::Image$new(
options=options,
name="emmPlot",
title="",
width=450,
height=400,
renderFun=".emmPlot",
visible="(emmPlots)",
clearWith=list(
"dep",
"blocks",
"weights",
"refLevels",
"ciEmm",
"ciWidthEmm",
"emmWeights")))
self$add(jmvcore::Table$new(
options=options,
name="emmTable",
title="",
visible="(emmTables)",
columns=list(),
clearWith=list(
"dep",
"blocks",
"weights",
"refLevels",
"ciWidthEmm",
"emmWeights")))}))$new(options=options)))}))$new(options=options)))
self$add(jmvcore::Output$new(
options=options,
name="predictOV",
title="Predicted values",
varTitle="Predicted values",
varDescription="Predicted values of linear regression",
measureType="continuous",
clearWith=list(
"dep",
"blocks",
"weights")))
self$add(jmvcore::Output$new(
options=options,
name="residsOV",
title="Residuals",
varTitle="Residuals",
varDescription="Residuals of linear regression",
measureType="continuous",
clearWith=list(
"dep",
"blocks",
"weights")))
self$add(jmvcore::Output$new(
options=options,
name="cooksOV",
title="Cook's Distance",
varTitle="Cook's distance",
varDescription="Cook's distance of linear regression",
measureType="continuous",
clearWith=list(
"dep",
"blocks",
"weights")))
self$add(jmvcore::Output$new(
options=options,
name="mahalOV",
title="Mahalanobis Distance",
varTitle="Mahalanobis",
varDescription="Mahalanobis distance (statistic and p-value) of the independent variables",
measureType="continuous",
clearWith=list(
"dep",
"blocks",
"weights")))}))
linRegBase <- if (requireNamespace("jmvcore", quietly=TRUE)) R6::R6Class(
"linRegBase",
inherit = jmvcore::Analysis,
public = list(
initialize = function(options, data=NULL, datasetId="", analysisId="", revision=0) {
super$initialize(
package = "jmv",
name = "linReg",
version = c(2,0,0),
options = options,
results = linRegResults$new(options=options),
data = data,
datasetId = datasetId,
analysisId = analysisId,
revision = revision,
pause = NULL,
completeWhenFilled = FALSE,
requiresMissings = FALSE,
weightsSupport = 'full')
}))
#' Linear Regression
#'
#' Linear regression is used to explore the relationship between a continuous
#' dependent variable, and one or more continuous and/or categorical
#' explanatory variables. Other statistical methods, such as ANOVA and ANCOVA,
#' are in reality just forms of linear regression.
#'
#'
#' @examples
#' data('Prestige', package='carData')
#'
#' linReg(data = Prestige, dep = income,
#' covs = vars(education, prestige, women),
#' blocks = list(list('education', 'prestige', 'women')))
#'
#' #
#' # LINEAR REGRESSION
#' #
#' # Model Fit Measures
#' # ---------------------------
#' # Model R R²
#' # ---------------------------
#' # 1 0.802 0.643
#' # ---------------------------
#' #
#' #
#' # MODEL SPECIFIC RESULTS
#' #
#' # MODEL 1
#' #
#' #
#' # Model Coefficients
#' # --------------------------------------------------------
#' # Predictor Estimate SE t p
#' # --------------------------------------------------------
#' # Intercept -253.8 1086.16 -0.234 0.816
#' # women -50.9 8.56 -5.948 < .001
#' # prestige 141.4 29.91 4.729 < .001
#' # education 177.2 187.63 0.944 0.347
#' # --------------------------------------------------------
#' #
#'
#' @param data the data as a data frame
#' @param dep the dependent variable from \code{data}, variable must be
#' numeric
#' @param covs the covariates from \code{data}
#' @param factors the fixed factors from \code{data}
#' @param weights the (optional) weights from \code{data} to be used in the
#' fitting process
#' @param blocks a list containing vectors of strings that name the predictors
#' that are added to the model. The elements are added to the model according
#' to their order in the list
#' @param refLevels a list of lists specifying reference levels of the
#' dependent variable and all the factors
#' @param intercept \code{'refLevel'} (default) or \code{'grandMean'}, coding
#' of the intercept. Either creates contrast so that the intercept represents
#' the reference level or the grand mean
#' @param r \code{TRUE} (default) or \code{FALSE}, provide the statistical
#' measure \code{R} for the models
#' @param r2 \code{TRUE} (default) or \code{FALSE}, provide the statistical
#' measure \code{R-squared} for the models
#' @param r2Adj \code{TRUE} or \code{FALSE} (default), provide the statistical
#' measure \code{adjusted R-squared} for the models
#' @param aic \code{TRUE} or \code{FALSE} (default), provide Aikaike's
#' Information Criterion (AIC) for the models
#' @param bic \code{TRUE} or \code{FALSE} (default), provide Bayesian
#' Information Criterion (BIC) for the models
#' @param rmse \code{TRUE} or \code{FALSE} (default), provide RMSE for the
#' models
#' @param modelTest \code{TRUE} (default) or \code{FALSE}, provide the model
#' comparison between the models and the NULL model
#' @param anova \code{TRUE} or \code{FALSE} (default), provide the omnibus
#' ANOVA test for the predictors
#' @param ci \code{TRUE} or \code{FALSE} (default), provide a confidence
#' interval for the model coefficients
#' @param ciWidth a number between 50 and 99.9 (default: 95) specifying the
#' confidence interval width
#' @param stdEst \code{TRUE} or \code{FALSE} (default), provide a standardized
#' estimate for the model coefficients
#' @param ciStdEst \code{TRUE} or \code{FALSE} (default), provide a confidence
#' interval for the model coefficient standardized estimates
#' @param ciWidthStdEst a number between 50 and 99.9 (default: 95) specifying
#' the confidence interval width
#' @param norm \code{TRUE} or \code{FALSE} (default), perform a Shapiro-Wilk
#' test on the residuals
#' @param qqPlot \code{TRUE} or \code{FALSE} (default), provide a Q-Q plot of
#' residuals
#' @param resPlots \code{TRUE} or \code{FALSE} (default), provide residual
#' plots where the dependent variable and each covariate is plotted against
#' the standardized residuals.
#' @param durbin \code{TRUE} or \code{FALSE} (default), provide results of the
#' Durbin- Watson test for autocorrelation
#' @param collin \code{TRUE} or \code{FALSE} (default), provide VIF and
#' tolerence collinearity statistics
#' @param cooks \code{TRUE} or \code{FALSE} (default), provide summary
#' statistics for the Cook's distance
#' @param mahal \code{TRUE} or \code{FALSE} (default), provide a summary table
#' reporting (significant) Mahalanobis distances
#' @param mahalp \code{'0.05'}, \code{'0.01'} or \code{'0.001'} (default),
#' p-threshold to be used for selecting entries in the summary table that
#' reports Mahalanobis distances that are significant at that p-threshold
#' @param emMeans a formula containing the terms to estimate marginal means
#' for, supports up to three variables per term
#' @param ciEmm \code{TRUE} (default) or \code{FALSE}, provide a confidence
#' interval for the estimated marginal means
#' @param ciWidthEmm a number between 50 and 99.9 (default: 95) specifying the
#' confidence interval width for the estimated marginal means
#' @param emmPlots \code{TRUE} (default) or \code{FALSE}, provide estimated
#' marginal means plots
#' @param emmTables \code{TRUE} or \code{FALSE} (default), provide estimated
#' marginal means tables
#' @param emmWeights \code{TRUE} (default) or \code{FALSE}, weigh each cell
#' equally or weigh them according to the cell frequency
#' @return A results object containing:
#' \tabular{llllll}{
#' \code{results$modelFit} \tab \tab \tab \tab \tab a table \cr
#' \code{results$modelComp} \tab \tab \tab \tab \tab a table \cr
#' \code{results$models} \tab \tab \tab \tab \tab an array of model specific results \cr
#' \code{results$predictOV} \tab \tab \tab \tab \tab an output \cr
#' \code{results$residsOV} \tab \tab \tab \tab \tab an output \cr
#' \code{results$cooksOV} \tab \tab \tab \tab \tab an output \cr
#' \code{results$mahalOV} \tab \tab \tab \tab \tab an output \cr
#' }
#'
#' Tables can be converted to data frames with \code{asDF} or \code{\link{as.data.frame}}. For example:
#'
#' \code{results$modelFit$asDF}
#'
#' \code{as.data.frame(results$modelFit)}
#'
#' @export
linReg <- function(
data,
dep,
covs = NULL,
factors = NULL,
weights = NULL,
blocks = list(
list()),
refLevels = NULL,
intercept = "refLevel",
r = TRUE,
r2 = TRUE,
r2Adj = FALSE,
aic = FALSE,
bic = FALSE,
rmse = FALSE,
modelTest = FALSE,
anova = FALSE,
ci = FALSE,
ciWidth = 95,
stdEst = FALSE,
ciStdEst = FALSE,
ciWidthStdEst = 95,
norm = FALSE,
qqPlot = FALSE,
resPlots = FALSE,
durbin = FALSE,
collin = FALSE,
cooks = FALSE,
mahal = FALSE,
mahalp = "0.001",
emMeans = list(
list()),
ciEmm = TRUE,
ciWidthEmm = 95,
emmPlots = TRUE,
emmTables = FALSE,
emmWeights = TRUE) {
if ( ! requireNamespace("jmvcore", quietly=TRUE))
stop("linReg requires jmvcore to be installed (restart may be required)")
if ( ! missing(dep)) dep <- jmvcore::resolveQuo(jmvcore::enquo(dep))
if ( ! missing(covs)) covs <- jmvcore::resolveQuo(jmvcore::enquo(covs))
if ( ! missing(factors)) factors <- jmvcore::resolveQuo(jmvcore::enquo(factors))
if ( ! missing(weights)) weights <- jmvcore::resolveQuo(jmvcore::enquo(weights))
if (missing(data))
data <- jmvcore::marshalData(
parent.frame(),
`if`( ! missing(dep), dep, NULL),
`if`( ! missing(covs), covs, NULL),
`if`( ! missing(factors), factors, NULL),
`if`( ! missing(weights), weights, NULL))
for (v in factors) if (v %in% names(data)) data[[v]] <- as.factor(data[[v]])
if (inherits(emMeans, "formula")) emMeans <- jmvcore::decomposeFormula(emMeans)
options <- linRegOptions$new(
dep = dep,
covs = covs,
factors = factors,
weights = weights,
blocks = blocks,
refLevels = refLevels,
intercept = intercept,
r = r,
r2 = r2,
r2Adj = r2Adj,
aic = aic,
bic = bic,
rmse = rmse,
modelTest = modelTest,
anova = anova,
ci = ci,
ciWidth = ciWidth,
stdEst = stdEst,
ciStdEst = ciStdEst,
ciWidthStdEst = ciWidthStdEst,
norm = norm,
qqPlot = qqPlot,
resPlots = resPlots,
durbin = durbin,
collin = collin,
cooks = cooks,
mahal = mahal,
mahalp = mahalp,
emMeans = emMeans,
ciEmm = ciEmm,
ciWidthEmm = ciWidthEmm,
emmPlots = emmPlots,
emmTables = emmTables,
emmWeights = emmWeights)
analysis <- linRegClass$new(
options = options,
data = data)
analysis$run()
analysis$results
}
jamovi/Rjamovi documentation built on Jan. 17, 2025, 10:29 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Defines functions linReg
# This file is automatically generated, you probably don't want to edit this
linRegOptions <- if (requireNamespace("jmvcore", quietly=TRUE)) R6::R6Class(
"linRegOptions",
inherit = jmvcore::Options,
public = list(
initialize = function(
dep = NULL,
covs = NULL,
factors = NULL,
weights = NULL,
blocks = list(
list()),
refLevels = NULL,
intercept = "refLevel",
r = TRUE,
r2 = TRUE,
r2Adj = FALSE,
aic = FALSE,
bic = FALSE,
rmse = FALSE,
modelTest = FALSE,
anova = FALSE,
ci = FALSE,
ciWidth = 95,
stdEst = FALSE,
ciStdEst = FALSE,
ciWidthStdEst = 95,
norm = FALSE,
qqPlot = FALSE,
resPlots = FALSE,
durbin = FALSE,
collin = FALSE,
cooks = FALSE,
mahal = FALSE,
mahalp = "0.001",
emMeans = list(
list()),
ciEmm = TRUE,
ciWidthEmm = 95,
emmPlots = TRUE,
emmTables = FALSE,
emmWeights = TRUE, ...) {
super$initialize(
package="jmv",
name="linReg",
requiresData=TRUE,
...)
private$..dep <- jmvcore::OptionVariable$new(
"dep",
dep,
suggested=list(
"continuous"),
permitted=list(
"numeric"))
private$..covs <- jmvcore::OptionVariables$new(
"covs",
covs,
suggested=list(
"continuous"),
permitted=list(
"numeric"),
default=NULL)
private$..factors <- jmvcore::OptionVariables$new(
"factors",
factors,
rejectUnusedLevels=TRUE,
suggested=list(
"nominal"),
permitted=list(
"factor"),
default=NULL)
private$..weights <- jmvcore::OptionVariable$new(
"weights",
weights,
suggested=list(
"continuous"),
permitted=list(
"numeric"),
default=NULL)
private$..blocks <- jmvcore::OptionArray$new(
"blocks",
blocks,
default=list(
list()),
template=jmvcore::OptionTerms$new(
"blocks",
NULL))
private$..refLevels <- jmvcore::OptionArray$new(
"refLevels",
refLevels,
items="(factors)",
default=NULL,
template=jmvcore::OptionGroup$new(
"refLevels",
NULL,
elements=list(
jmvcore::OptionVariable$new(
"var",
NULL,
content="$key"),
jmvcore::OptionLevel$new(
"ref",
NULL))))
private$..intercept <- jmvcore::OptionList$new(
"intercept",
intercept,
options=list(
"refLevel",
"grandMean"),
default="refLevel")
private$..r <- jmvcore::OptionBool$new(
"r",
r,
default=TRUE)
private$..r2 <- jmvcore::OptionBool$new(
"r2",
r2,
default=TRUE)
private$..r2Adj <- jmvcore::OptionBool$new(
"r2Adj",
r2Adj,
default=FALSE)
private$..aic <- jmvcore::OptionBool$new(
"aic",
aic,
default=FALSE)
private$..bic <- jmvcore::OptionBool$new(
"bic",
bic,
default=FALSE)
private$..rmse <- jmvcore::OptionBool$new(
"rmse",
rmse,
default=FALSE)
private$..modelTest <- jmvcore::OptionBool$new(
"modelTest",
modelTest,
default=FALSE)
private$..anova <- jmvcore::OptionBool$new(
"anova",
anova,
default=FALSE)
private$..ci <- jmvcore::OptionBool$new(
"ci",
ci,
default=FALSE)
private$..ciWidth <- jmvcore::OptionNumber$new(
"ciWidth",
ciWidth,
min=50,
max=99.9,
default=95)
private$..stdEst <- jmvcore::OptionBool$new(
"stdEst",
stdEst,
default=FALSE)
private$..ciStdEst <- jmvcore::OptionBool$new(
"ciStdEst",
ciStdEst,
default=FALSE)
private$..ciWidthStdEst <- jmvcore::OptionNumber$new(
"ciWidthStdEst",
ciWidthStdEst,
min=50,
max=99.9,
default=95)
private$..norm <- jmvcore::OptionBool$new(
"norm",
norm,
default=FALSE)
private$..qqPlot <- jmvcore::OptionBool$new(
"qqPlot",
qqPlot,
default=FALSE)
private$..resPlots <- jmvcore::OptionBool$new(
"resPlots",
resPlots,
default=FALSE)
private$..durbin <- jmvcore::OptionBool$new(
"durbin",
durbin,
default=FALSE)
private$..collin <- jmvcore::OptionBool$new(
"collin",
collin,
default=FALSE)
private$..cooks <- jmvcore::OptionBool$new(
"cooks",
cooks,
default=FALSE)
private$..mahal <- jmvcore::OptionBool$new(
"mahal",
mahal,
default=FALSE)
private$..mahalp <- jmvcore::OptionList$new(
"mahalp",
mahalp,
options=list(
"0.05",
"0.01",
"0.001"),
default="0.001")
private$..emMeans <- jmvcore::OptionArray$new(
"emMeans",
emMeans,
default=list(
list()),
template=jmvcore::OptionVariables$new(
"emMeans",
NULL))
private$..ciEmm <- jmvcore::OptionBool$new(
"ciEmm",
ciEmm,
default=TRUE)
private$..ciWidthEmm <- jmvcore::OptionNumber$new(
"ciWidthEmm",
ciWidthEmm,
min=50,
max=99.9,
default=95)
private$..emmPlots <- jmvcore::OptionBool$new(
"emmPlots",
emmPlots,
default=TRUE)
private$..emmTables <- jmvcore::OptionBool$new(
"emmTables",
emmTables,
default=FALSE)
private$..emmWeights <- jmvcore::OptionBool$new(
"emmWeights",
emmWeights,
default=TRUE)
private$..predictOV <- jmvcore::OptionOutput$new(
"predictOV")
private$..residsOV <- jmvcore::OptionOutput$new(
"residsOV")
private$..cooksOV <- jmvcore::OptionOutput$new(
"cooksOV")
private$..mahalOV <- jmvcore::OptionOutput$new(
"mahalOV")
self$.addOption(private$..dep)
self$.addOption(private$..covs)
self$.addOption(private$..factors)
self$.addOption(private$..weights)
self$.addOption(private$..blocks)
self$.addOption(private$..refLevels)
self$.addOption(private$..intercept)
self$.addOption(private$..r)
self$.addOption(private$..r2)
self$.addOption(private$..r2Adj)
self$.addOption(private$..aic)
self$.addOption(private$..bic)
self$.addOption(private$..rmse)
self$.addOption(private$..modelTest)
self$.addOption(private$..anova)
self$.addOption(private$..ci)
self$.addOption(private$..ciWidth)
self$.addOption(private$..stdEst)
self$.addOption(private$..ciStdEst)
self$.addOption(private$..ciWidthStdEst)
self$.addOption(private$..norm)
self$.addOption(private$..qqPlot)
self$.addOption(private$..resPlots)
self$.addOption(private$..durbin)
self$.addOption(private$..collin)
self$.addOption(private$..cooks)
self$.addOption(private$..mahal)
self$.addOption(private$..mahalp)
self$.addOption(private$..emMeans)
self$.addOption(private$..ciEmm)
self$.addOption(private$..ciWidthEmm)
self$.addOption(private$..emmPlots)
self$.addOption(private$..emmTables)
self$.addOption(private$..emmWeights)
self$.addOption(private$..predictOV)
self$.addOption(private$..residsOV)
self$.addOption(private$..cooksOV)
self$.addOption(private$..mahalOV)
}),
active = list(
dep = function() private$..dep$value,
covs = function() private$..covs$value,
factors = function() private$..factors$value,
weights = function() private$..weights$value,
blocks = function() private$..blocks$value,
refLevels = function() private$..refLevels$value,
intercept = function() private$..intercept$value,
r = function() private$..r$value,
r2 = function() private$..r2$value,
r2Adj = function() private$..r2Adj$value,
aic = function() private$..aic$value,
bic = function() private$..bic$value,
rmse = function() private$..rmse$value,
modelTest = function() private$..modelTest$value,
anova = function() private$..anova$value,
ci = function() private$..ci$value,
ciWidth = function() private$..ciWidth$value,
stdEst = function() private$..stdEst$value,
ciStdEst = function() private$..ciStdEst$value,
ciWidthStdEst = function() private$..ciWidthStdEst$value,
norm = function() private$..norm$value,
qqPlot = function() private$..qqPlot$value,
resPlots = function() private$..resPlots$value,
durbin = function() private$..durbin$value,
collin = function() private$..collin$value,
cooks = function() private$..cooks$value,
mahal = function() private$..mahal$value,
mahalp = function() private$..mahalp$value,
emMeans = function() private$..emMeans$value,
ciEmm = function() private$..ciEmm$value,
ciWidthEmm = function() private$..ciWidthEmm$value,
emmPlots = function() private$..emmPlots$value,
emmTables = function() private$..emmTables$value,
emmWeights = function() private$..emmWeights$value,
predictOV = function() private$..predictOV$value,
residsOV = function() private$..residsOV$value,
cooksOV = function() private$..cooksOV$value,
mahalOV = function() private$..mahalOV$value),
private = list(
..dep = NA,
..covs = NA,
..factors = NA,
..weights = NA,
..blocks = NA,
..refLevels = NA,
..intercept = NA,
..r = NA,
..r2 = NA,
..r2Adj = NA,
..aic = NA,
..bic = NA,
..rmse = NA,
..modelTest = NA,
..anova = NA,
..ci = NA,
..ciWidth = NA,
..stdEst = NA,
..ciStdEst = NA,
..ciWidthStdEst = NA,
..norm = NA,
..qqPlot = NA,
..resPlots = NA,
..durbin = NA,
..collin = NA,
..cooks = NA,
..mahal = NA,
..mahalp = NA,
..emMeans = NA,
..ciEmm = NA,
..ciWidthEmm = NA,
..emmPlots = NA,
..emmTables = NA,
..emmWeights = NA,
..predictOV = NA,
..residsOV = NA,
..cooksOV = NA,
..mahalOV = NA)
)
linRegResults <- if (requireNamespace("jmvcore", quietly=TRUE)) R6::R6Class(
"linRegResults",
inherit = jmvcore::Group,
active = list(
modelFit = function() private$.items[["modelFit"]],
modelComp = function() private$.items[["modelComp"]],
models = function() private$.items[["models"]],
predictOV = function() private$.items[["predictOV"]],
residsOV = function() private$.items[["residsOV"]],
cooksOV = function() private$.items[["cooksOV"]],
mahalOV = function() private$.items[["mahalOV"]]),
private = list(),
public=list(
initialize=function(options) {
super$initialize(
options=options,
name="",
title="Linear Regression")
self$add(jmvcore::Table$new(
options=options,
name="modelFit",
title="Model Fit Measures",
clearWith=list(
"dep",
"blocks",
"weights"),
visible="(r || r2 || r2Adj || aic || bic || rmse || modelTest)",
columns=list(
list(
`name`="model",
`title`="Model",
`type`="text"),
list(
`name`="r",
`title`="R",
`type`="number",
`visible`="(r)"),
list(
`name`="r2",
`title`="R\u00B2",
`type`="number",
`visible`="(r2)"),
list(
`name`="r2Adj",
`title`="Adjusted R\u00B2",
`type`="number",
`visible`="(r2Adj)"),
list(
`name`="aic",
`title`="AIC",
`type`="number",
`visible`="(aic)"),
list(
`name`="bic",
`title`="BIC",
`type`="number",
`visible`="(bic)"),
list(
`name`="rmse",
`title`="RMSE",
`type`="number",
`visible`="(rmse)"),
list(
`name`="f",
`title`="F",
`type`="number",
`superTitle`="Overall Model Test",
`visible`="(modelTest)"),
list(
`name`="df1",
`title`="df1",
`type`="integer",
`superTitle`="Overall Model Test",
`visible`="(modelTest)"),
list(
`name`="df2",
`title`="df2",
`type`="integer",
`superTitle`="Overall Model Test",
`visible`="(modelTest)"),
list(
`name`="p",
`title`="p",
`type`="number",
`format`="zto,pvalue",
`superTitle`="Overall Model Test",
`visible`="(modelTest)"))))
self$add(jmvcore::Table$new(
options=options,
name="modelComp",
title="Model Comparisons",
clearWith=list(
"dep",
"blocks",
"weights"),
columns=list(
list(
`name`="model1",
`title`="Model",
`content`=".",
`type`="text",
`superTitle`="Comparison"),
list(
`name`="sep",
`title`="",
`content`="-",
`type`="text",
`format`="narrow",
`superTitle`="Comparison"),
list(
`name`="model2",
`title`="Model",
`content`=".",
`type`="text",
`superTitle`="Comparison"),
list(
`name`="r2",
`title`="\u0394R\u00B2",
`type`="number"),
list(
`name`="f",
`title`="F",
`type`="number"),
list(
`name`="df1",
`title`="df1",
`type`="integer"),
list(
`name`="df2",
`title`="df2",
`type`="integer"),
list(
`name`="p",
`title`="p",
`type`="number",
`format`="zto,pvalue"))))
self$add(jmvcore::Array$new(
options=options,
name="models",
title="Model Specific Results",
layout="listSelect",
hideHeadingOnlyChild=TRUE,
clearWith=list(
"blocks"),
template=R6::R6Class(
inherit = jmvcore::Group,
active = list(
anova = function() private$.items[["anova"]],
coef = function() private$.items[["coef"]],
dataSummary = function() private$.items[["dataSummary"]],
assump = function() private$.items[["assump"]],
emm = function() private$.items[["emm"]]),
private = list(),
public=list(
initialize=function(options) {
super$initialize(
options=options,
name="undefined",
title="")
self$add(jmvcore::Table$new(
options=options,
name="anova",
title="Omnibus ANOVA Test",
visible="(anova)",
refs="car",
clearWith=list(
"dep",
"blocks",
"weights"),
columns=list(
list(
`name`="term",
`title`="",
`type`="text"),
list(
`name`="ss",
`title`="Sum of Squares",
`type`="number"),
list(
`name`="df",
`title`="df",
`type`="integer"),
list(
`name`="ms",
`title`="Mean Square",
`type`="number"),
list(
`name`="F",
`title`="F",
`type`="number"),
list(
`name`="p",
`title`="p",
`type`="number",
`format`="zto,pvalue"))))
self$add(jmvcore::Table$new(
options=options,
name="coef",
title="`Model Coefficients - ${ dep }`",
clearWith=list(
"dep",
"blocks",
"refLevels",
"ciWidth",
"ciWidthStdEst",
"intercept",
"weights"),
columns=list(
list(
`name`="term",
`title`="Predictor",
`type`="text"),
list(
`name`="est",
`title`="Estimate",
`type`="number"),
list(
`name`="se",
`title`="SE",
`type`="number"),
list(
`name`="lower",
`title`="Lower",
`type`="number",
`visible`="(ci)"),
list(
`name`="upper",
`title`="Upper",
`type`="number",
`visible`="(ci)"),
list(
`name`="t",
`title`="t",
`type`="number"),
list(
`name`="p",
`title`="p",
`type`="number",
`format`="zto,pvalue"),
list(
`name`="stdEst",
`title`="Stand. Estimate",
`type`="number",
`visible`="(stdEst)"),
list(
`name`="stdEstLower",
`title`="Lower",
`type`="number",
`visible`="(ciStdEst && stdEst)"),
list(
`name`="stdEstUpper",
`title`="Upper",
`type`="number",
`visible`="(ciStdEst && stdEst)"))))
self$add(R6::R6Class(
inherit = jmvcore::Group,
active = list(
cooks = function() private$.items[["cooks"]],
mahal = function() private$.items[["mahal"]]),
private = list(),
public=list(
initialize=function(options) {
super$initialize(
options=options,
name="dataSummary",
title="Data Summary")
self$add(jmvcore::Table$new(
options=options,
name="cooks",
title="Cook's Distance",
rows=1,
visible="(cooks)",
clearWith=list(
"dep",
"blocks",
"weights"),
columns=list(
list(
`name`="mean",
`title`="Mean",
`type`="number"),
list(
`name`="median",
`title`="Median",
`type`="number"),
list(
`name`="sd",
`title`="SD",
`type`="number"),
list(
`name`="min",
`title`="Min",
`type`="number",
`superTitle`="Range"),
list(
`name`="max",
`title`="Max",
`type`="number",
`superTitle`="Range"))))
self$add(jmvcore::Table$new(
options=options,
name="mahal",
title="Mahalanobis Distance",
rows=1,
visible="(mahal)",
clearWith=list(
"dep",
"blocks",
"weights"),
columns=list(
list(
`name`="mean",
`title`="Mean",
`type`="number"),
list(
`name`="median",
`title`="Median",
`type`="number"),
list(
`name`="sd",
`title`="SD",
`type`="number"),
list(
`name`="min",
`title`="Min",
`type`="number",
`superTitle`="Range"),
list(
`name`="max",
`title`="Max",
`type`="number",
`superTitle`="Range"),
list(
`name`="excRow",
`title`="Participants (rows) below p-threshold",
`type`="text"))))}))$new(options=options))
self$add(R6::R6Class(
inherit = jmvcore::Group,
active = list(
durbin = function() private$.items[["durbin"]],
collin = function() private$.items[["collin"]],
norm = function() private$.items[["norm"]],
qqPlot = function() private$.items[["qqPlot"]],
resPlots = function() private$.items[["resPlots"]]),
private = list(),
public=list(
initialize=function(options) {
super$initialize(
options=options,
name="assump",
title="Assumption Checks")
self$add(jmvcore::Table$new(
options=options,
name="durbin",
title="Durbin\u2013Watson Test for Autocorrelation",
rows=1,
visible="(durbin)",
refs="car",
clearWith=list(
"dep",
"blocks",
"weights"),
columns=list(
list(
`name`="autoCor",
`title`="Autocorrelation",
`type`="number"),
list(
`name`="dw",
`title`="DW Statistic",
`type`="number"),
list(
`name`="p",
`title`="p",
`type`="number",
`format`="zto,pvalue"))))
self$add(jmvcore::Table$new(
options=options,
name="collin",
title="Collinearity Statistics",
visible="(collin)",
refs="car",
clearWith=list(
"dep",
"blocks",
"weights"),
columns=list(
list(
`name`="term",
`title`="",
`type`="text"),
list(
`name`="vif",
`title`="VIF",
`type`="number"),
list(
`name`="tol",
`title`="Tolerance",
`type`="number"))))
self$add(jmvcore::Table$new(
options=options,
name="norm",
title="Normality Test (Shapiro-Wilk)",
visible="(norm)",
rows=1,
clearWith=list(
"dep",
"blocks",
"weights"),
columns=list(
list(
`name`="t[sw]",
`title`="",
`type`="text",
`content`="Shapiro-Wilk",
`visible`=FALSE),
list(
`name`="s[sw]",
`title`="Statistic"),
list(
`name`="p[sw]",
`title`="p",
`format`="zto,pvalue"))))
self$add(jmvcore::Image$new(
options=options,
name="qqPlot",
title="Q-Q Plot",
width=450,
height=400,
renderFun=".qqPlot",
visible="(qqPlot)",
requiresData=TRUE,
clearWith=list(
"dep",
"blocks",
"weights")))
self$add(jmvcore::Array$new(
options=options,
name="resPlots",
title="Residuals Plots",
visible="(resPlots)",
template=jmvcore::Image$new(
options=options,
renderFun=".resPlot",
requiresData=TRUE,
clearWith=list(
"dep",
"blocks",
"weights"))))}))$new(options=options))
self$add(jmvcore::Array$new(
options=options,
name="emm",
title="Estimated Marginal Means",
refs="emmeans",
clearWith=list(
"dep",
"blocks",
"weights",
"emMeans"),
template=R6::R6Class(
inherit = jmvcore::Group,
active = list(
emmPlot = function() private$.items[["emmPlot"]],
emmTable = function() private$.items[["emmTable"]]),
private = list(),
public=list(
initialize=function(options) {
super$initialize(
options=options,
name="undefined",
title="")
self$add(jmvcore::Image$new(
options=options,
name="emmPlot",
title="",
width=450,
height=400,
renderFun=".emmPlot",
visible="(emmPlots)",
clearWith=list(
"dep",
"blocks",
"weights",
"refLevels",
"ciEmm",
"ciWidthEmm",
"emmWeights")))
self$add(jmvcore::Table$new(
options=options,
name="emmTable",
title="",
visible="(emmTables)",
columns=list(),
clearWith=list(
"dep",
"blocks",
"weights",
"refLevels",
"ciWidthEmm",
"emmWeights")))}))$new(options=options)))}))$new(options=options)))
self$add(jmvcore::Output$new(
options=options,
name="predictOV",
title="Predicted values",
varTitle="Predicted values",
varDescription="Predicted values of linear regression",
measureType="continuous",
clearWith=list(
"dep",
"blocks",
"weights")))
self$add(jmvcore::Output$new(
options=options,
name="residsOV",
title="Residuals",
varTitle="Residuals",
varDescription="Residuals of linear regression",
measureType="continuous",
clearWith=list(
"dep",
"blocks",
"weights")))
self$add(jmvcore::Output$new(
options=options,
name="cooksOV",
title="Cook's Distance",
varTitle="Cook's distance",
varDescription="Cook's distance of linear regression",
measureType="continuous",
clearWith=list(
"dep",
"blocks",
"weights")))
self$add(jmvcore::Output$new(
options=options,
name="mahalOV",
title="Mahalanobis Distance",
varTitle="Mahalanobis",
varDescription="Mahalanobis distance (statistic and p-value) of the independent variables",
measureType="continuous",
clearWith=list(
"dep",
"blocks",
"weights")))}))
linRegBase <- if (requireNamespace("jmvcore", quietly=TRUE)) R6::R6Class(
"linRegBase",
inherit = jmvcore::Analysis,
public = list(
initialize = function(options, data=NULL, datasetId="", analysisId="", revision=0) {
super$initialize(
package = "jmv",
name = "linReg",
version = c(2,0,0),
options = options,
results = linRegResults$new(options=options),
data = data,
datasetId = datasetId,
analysisId = analysisId,
revision = revision,
pause = NULL,
completeWhenFilled = FALSE,
requiresMissings = FALSE,
weightsSupport = 'full')
}))
#' Linear Regression
#'
#' Linear regression is used to explore the relationship between a continuous
#' dependent variable, and one or more continuous and/or categorical
#' explanatory variables. Other statistical methods, such as ANOVA and ANCOVA,
#' are in reality just forms of linear regression.
#'
#'
#' @examples
#' data('Prestige', package='carData')
#'
#' linReg(data = Prestige, dep = income,
#' covs = vars(education, prestige, women),
#' blocks = list(list('education', 'prestige', 'women')))
#'
#' #
#' # LINEAR REGRESSION
#' #
#' # Model Fit Measures
#' # ---------------------------
#' # Model R R²
#' # ---------------------------
#' # 1 0.802 0.643
#' # ---------------------------
#' #
#' #
#' # MODEL SPECIFIC RESULTS
#' #
#' # MODEL 1
#' #
#' #
#' # Model Coefficients
#' # --------------------------------------------------------
#' # Predictor Estimate SE t p
#' # --------------------------------------------------------
#' # Intercept -253.8 1086.16 -0.234 0.816
#' # women -50.9 8.56 -5.948 < .001
#' # prestige 141.4 29.91 4.729 < .001
#' # education 177.2 187.63 0.944 0.347
#' # --------------------------------------------------------
#' #
#'
#' @param data the data as a data frame
#' @param dep the dependent variable from \code{data}, variable must be
#' numeric
#' @param covs the covariates from \code{data}
#' @param factors the fixed factors from \code{data}
#' @param weights the (optional) weights from \code{data} to be used in the
#' fitting process
#' @param blocks a list containing vectors of strings that name the predictors
#' that are added to the model. The elements are added to the model according
#' to their order in the list
#' @param refLevels a list of lists specifying reference levels of the
#' dependent variable and all the factors
#' @param intercept \code{'refLevel'} (default) or \code{'grandMean'}, coding
#' of the intercept. Either creates contrast so that the intercept represents
#' the reference level or the grand mean
#' @param r \code{TRUE} (default) or \code{FALSE}, provide the statistical
#' measure \code{R} for the models
#' @param r2 \code{TRUE} (default) or \code{FALSE}, provide the statistical
#' measure \code{R-squared} for the models
#' @param r2Adj \code{TRUE} or \code{FALSE} (default), provide the statistical
#' measure \code{adjusted R-squared} for the models
#' @param aic \code{TRUE} or \code{FALSE} (default), provide Aikaike's
#' Information Criterion (AIC) for the models
#' @param bic \code{TRUE} or \code{FALSE} (default), provide Bayesian
#' Information Criterion (BIC) for the models
#' @param rmse \code{TRUE} or \code{FALSE} (default), provide RMSE for the
#' models
#' @param modelTest \code{TRUE} (default) or \code{FALSE}, provide the model
#' comparison between the models and the NULL model
#' @param anova \code{TRUE} or \code{FALSE} (default), provide the omnibus
#' ANOVA test for the predictors
#' @param ci \code{TRUE} or \code{FALSE} (default), provide a confidence
#' interval for the model coefficients
#' @param ciWidth a number between 50 and 99.9 (default: 95) specifying the
#' confidence interval width
#' @param stdEst \code{TRUE} or \code{FALSE} (default), provide a standardized
#' estimate for the model coefficients
#' @param ciStdEst \code{TRUE} or \code{FALSE} (default), provide a confidence
#' interval for the model coefficient standardized estimates
#' @param ciWidthStdEst a number between 50 and 99.9 (default: 95) specifying
#' the confidence interval width
#' @param norm \code{TRUE} or \code{FALSE} (default), perform a Shapiro-Wilk
#' test on the residuals
#' @param qqPlot \code{TRUE} or \code{FALSE} (default), provide a Q-Q plot of
#' residuals
#' @param resPlots \code{TRUE} or \code{FALSE} (default), provide residual
#' plots where the dependent variable and each covariate is plotted against
#' the standardized residuals.
#' @param durbin \code{TRUE} or \code{FALSE} (default), provide results of the
#' Durbin- Watson test for autocorrelation
#' @param collin \code{TRUE} or \code{FALSE} (default), provide VIF and
#' tolerence collinearity statistics
#' @param cooks \code{TRUE} or \code{FALSE} (default), provide summary
#' statistics for the Cook's distance
#' @param mahal \code{TRUE} or \code{FALSE} (default), provide a summary table
#' reporting (significant) Mahalanobis distances
#' @param mahalp \code{'0.05'}, \code{'0.01'} or \code{'0.001'} (default),
#' p-threshold to be used for selecting entries in the summary table that
#' reports Mahalanobis distances that are significant at that p-threshold
#' @param emMeans a formula containing the terms to estimate marginal means
#' for, supports up to three variables per term
#' @param ciEmm \code{TRUE} (default) or \code{FALSE}, provide a confidence
#' interval for the estimated marginal means
#' @param ciWidthEmm a number between 50 and 99.9 (default: 95) specifying the
#' confidence interval width for the estimated marginal means
#' @param emmPlots \code{TRUE} (default) or \code{FALSE}, provide estimated
#' marginal means plots
#' @param emmTables \code{TRUE} or \code{FALSE} (default), provide estimated
#' marginal means tables
#' @param emmWeights \code{TRUE} (default) or \code{FALSE}, weigh each cell
#' equally or weigh them according to the cell frequency
#' @return A results object containing:
#' \tabular{llllll}{
#' \code{results$modelFit} \tab \tab \tab \tab \tab a table \cr
#' \code{results$modelComp} \tab \tab \tab \tab \tab a table \cr
#' \code{results$models} \tab \tab \tab \tab \tab an array of model specific results \cr
#' \code{results$predictOV} \tab \tab \tab \tab \tab an output \cr
#' \code{results$residsOV} \tab \tab \tab \tab \tab an output \cr
#' \code{results$cooksOV} \tab \tab \tab \tab \tab an output \cr
#' \code{results$mahalOV} \tab \tab \tab \tab \tab an output \cr
#' }
#'
#' Tables can be converted to data frames with \code{asDF} or \code{\link{as.data.frame}}. For example:
#'
#' \code{results$modelFit$asDF}
#'
#' \code{as.data.frame(results$modelFit)}
#'
#' @export
linReg <- function(
data,
dep,
covs = NULL,
factors = NULL,
weights = NULL,
blocks = list(
list()),
refLevels = NULL,
intercept = "refLevel",
r = TRUE,
r2 = TRUE,
r2Adj = FALSE,
aic = FALSE,
bic = FALSE,
rmse = FALSE,
modelTest = FALSE,
anova = FALSE,
ci = FALSE,
ciWidth = 95,
stdEst = FALSE,
ciStdEst = FALSE,
ciWidthStdEst = 95,
norm = FALSE,
qqPlot = FALSE,
resPlots = FALSE,
durbin = FALSE,
collin = FALSE,
cooks = FALSE,
mahal = FALSE,
mahalp = "0.001",
emMeans = list(
list()),
ciEmm = TRUE,
ciWidthEmm = 95,
emmPlots = TRUE,
emmTables = FALSE,
emmWeights = TRUE) {
if ( ! requireNamespace("jmvcore", quietly=TRUE))
stop("linReg requires jmvcore to be installed (restart may be required)")
if ( ! missing(dep)) dep <- jmvcore::resolveQuo(jmvcore::enquo(dep))
if ( ! missing(covs)) covs <- jmvcore::resolveQuo(jmvcore::enquo(covs))
if ( ! missing(factors)) factors <- jmvcore::resolveQuo(jmvcore::enquo(factors))
if ( ! missing(weights)) weights <- jmvcore::resolveQuo(jmvcore::enquo(weights))
if (missing(data))
data <- jmvcore::marshalData(
parent.frame(),
`if`( ! missing(dep), dep, NULL),
`if`( ! missing(covs), covs, NULL),
`if`( ! missing(factors), factors, NULL),
`if`( ! missing(weights), weights, NULL))
for (v in factors) if (v %in% names(data)) data[[v]] <- as.factor(data[[v]])
if (inherits(emMeans, "formula")) emMeans <- jmvcore::decomposeFormula(emMeans)
options <- linRegOptions$new(
dep = dep,
covs = covs,
factors = factors,
weights = weights,
blocks = blocks,
refLevels = refLevels,
intercept = intercept,
r = r,
r2 = r2,
r2Adj = r2Adj,
aic = aic,
bic = bic,
rmse = rmse,
modelTest = modelTest,
anova = anova,
ci = ci,
ciWidth = ciWidth,
stdEst = stdEst,
ciStdEst = ciStdEst,
ciWidthStdEst = ciWidthStdEst,
norm = norm,
qqPlot = qqPlot,
resPlots = resPlots,
durbin = durbin,
collin = collin,
cooks = cooks,
mahal = mahal,
mahalp = mahalp,
emMeans = emMeans,
ciEmm = ciEmm,
ciWidthEmm = ciWidthEmm,
emmPlots = emmPlots,
emmTables = emmTables,
emmWeights = emmWeights)
analysis <- linRegClass$new(
options = options,
data = data)
analysis$run()
analysis$results
}
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.