Robust Regression (Least Trimmed Squares) Dialog Functions

DESCRIPTION:

These functions are used by the Robust Regression dialog. menuLtsreg calls tabSummary.lts and tabPlot.lts if summary and plotting results are requested.

USAGE:

menuLtsreg(formula, data, weights, subset, na.omit.p=T, 
           quan=NULL, print.short.p=F, print.long.p=T, 
           save.name=NULL, save.fit.p=F, save.resid.p=F, 
           save.weights.p=F, plotResidVsFit.p=F, 
           plotResidVsIndex.p=F, plotQQ.p=F, plotResidVsDist.p=F, 
           id.n=3) 
tabSummary.lts(ltsobj, print.short.p=F, print.long.p=T, 
           save.name=NULL, save.fit.p=F, save.resid.p=F, 
           save.weights.p=F) 
tabPlot.lts(ltsobj, plotResidVsFit.p=F, plotResidVsIndex.p=F, 
           plotQQ.p=F, plotResidVsDist.p=F, id.n=3, ...) 

REQUIRED ARGUMENTS:

formula
a formula object, with the response on the left of a `~' operator, and the terms, separated by + operators, on the right.
ltsobj
n object that inherits from class lts.

OPTIONAL ARGUMENTS:

data
a data.frame in which to interpret the variables named in the formula, or in the subset and the weights argument. If this is missing, then the variables in the formula should be on the search list.
weights
vector of observation weights; if supplied, the algorithm fits to minimize the sum of the weights multiplied into the squared residuals. The length of weights must be the same as the number of observations. The weights must be nonnegative and it is strongly recommended that they be strictly positive, since zero weights are ambiguous, compared to use of the subset argument.
subset
expression saying which subset of the rows of the data should be used in the fit. This can be a logical vector (which is replicated to have length equal to the number of observations), or a numeric vector indicating which observation numbers are to be included, or a character vector of the row names to be included. All observations are included by default.
na.omit.p
if TRUE, then any observation with missing values are removed from the analysis. If FALSE and there are missing values then the function will exit with a message that missing values are not allowed. If na.omit.p is TRUE then na.action is set to na.omit in the call to ltsreg. If na.omit.p is FALSE then na.action is set to na.fail in the call to ltsreg.
quan
the number of squared residuals whose sum will be minimized. This defaults to floor((n+p+1)/2) where n is the number of observations and p s the number of predictors in the model.
print.short.p
if TRUE, a short summary of the linear regression is printed. This output is from the function print.lts.
print.long.p
if TRUE, a long summary of the linear regression is printed. This output is from the function summary.lts.
save.name
a character string for the name of the data frame to save the fit and residuals in. If data frame with this name already exists in database 1 and it has the appropriate number of rows then the saved values will be appended to the data frame. If the object already exist in database 1 and it is not a data frame or it does not have the appropriate number of rows then a new name is created by appending a number to save.name and the results are saved in the data frame with the new name.
save.fit.p
if TRUE, the fitted values from the regression are saved in the data frame save.name.
save.resid.p
if TRUE, the residuals from the regression are saved in the data frame save.name.
save.weights.p
if TRUE, weights with a value of 1 for observations with reasonably small residuals and a value of 0 for observations with large residuals are saved. These weights can latter be used in an ordinary least squares regression.
plotResidVsFit.p
f TRUE, a plot of the standardized residuals versus the fitted values is created.
plotResidVsIndex.p
if TRUE, a plot of the standardized residuals versus the index of the observations is created.
plotQQ.p
if TRUE, a Normal quantile-quantile plot of the LTS residuals is drawn.
plotResidVsDist.p
if TRUE, a plot of the LTS residuals vs Robust Distances of x-rows is created.
id.n
he number of extreme points that will be identified on the plots. The row names from the model's data frame will be used to identify the points.

VALUE:

invisibly returns an object of class lts. See the lts.object help file for details.

SIDE EFFECTS:

Printed output will be displayed if requested. Plots will be drawn if requested. The objects save.name and predobj.name will be created or appended to if fitted values, residuals or predictions are saved

SEE ALSO:

, , .