mInfluence: Interactive applet to explore influential points in linear...

In mosaicManip: Project MOSAIC (mosaic-web.org) manipulate examples.

Description

Displays data and a menu of modeling functions. Finds a linear combination of the selected modeling functions, with the user setting nonlinear parameters manually.

Usage

mInfluence(expr, data, groups=NULL, 
			col.fitted = 'gray30', col.influence='red', ...)

Arguments

expr	A formula object giving the dependent and independent variables to use.
data	A data frame containing the values for the variables identified in expr.
groups	A variable or expression to be evaluated in data, expected to act as a grouping variable used to used to distinguish different groups by varying graphical parameters like color and plot symbol.
col.fitted, col.influence	colors to use for indicating influential points and fitted points. Set to 'transparent' to hid these elements.
...	Additional arguments passed to xyplot to control the plotting.

Details

The linear model described by the expr term is found and fitted values are plotted against the data. Using the Dep. variable to plot picker the user may choose which variable to plot on the x axis. The user can add a new point to the data set by checking the Add new point checkbox. Using the x and y sliders below it, the user can change the position of the new point. The sliders' units are standard deviations from the mean. The type of influence picker is currently nonfunctional.

Author(s)

Andrew Rich ([e-mail]), Daniel Kaplan ([e-mail]), Randall Pruim ([e-mail])

Examples

if(require(manipulate)) {
data(trees); fetchData(CPS)
mInfluence(Girth~Height, trees)
mInfluence(wage~educ+exper+race+sex+age+sector, data=CPS)
}

mosaicManip documentation built on May 2, 2019, 5:48 p.m.