RSDA (version 1.1)

sym.lm.bi: Symbolic Linear Regression for two variables

Description

The function build a symbolic regression for two interval or continuos variables.

Usage

sym.lm.bi(sym.var.x, sym.var.y, method = c("mid-points", "tops", "inf-sup", 
                 "billard"))

Arguments

sym.var.x

The firth symbolic variable.

sym.var.y

The second symbolic variable.

method

The thirth symbolic variable.

Value

• This function return a regression structure as folllows: $Intercept [1] 38.64236 $Beta1 [1] 0.3081313

References

Billard L. and Diday E. (2006). Symbolic data analysis: Conceptual statistics and data mining. Wiley, Chichester. Rodriguez, O. (2000). Classification et Modeles Lineaires en Analyse des Donnees Symboliques. Ph.D. Thesis, Paris IX-Dauphine University.

Examples

data(example3)
sym.data<-example3
lm.mod<-sym.lm.bi(sym.var(sym.data,1),sym.var(sym.data,4))
sym.scatterplot(sym.var(sym.data,1),sym.var(sym.data,4),col='blue',
                main='Linear Regression')
abline(lm.mod,lwd=3)

lm.mod<-sym.lm.bi(sym.var(sym.data,2),sym.var(sym.data,6))
sym.scatterplot(sym.var(sym.data,2),sym.var(sym.data,6),
col='blue',main='Linear Regression')
abline(lm.mod,lwd=3)

data(lynne1)
lm.mod<-sym.lm.bi(sym.var(lynne1,2),sym.var(lynne1,1))
sym.scatterplot(sym.var(lynne1,2),sym.var(lynne1,1),labels=TRUE, 
col='red',main='Linear Regression')
abline(lm.mod,lwd=3,col='blue')

lm.mod<-sym.lm.bi(sym.var(lynne1,2),sym.var(lynne1,1),method='inf-sup')
sym.scatterplot(sym.var(lynne1,2),sym.var(lynne1,1),labels=TRUE,
col='red',main='Linear Regression')
abline(lm.mod$inf,lwd=3,col='blue')
abline(lm.mod$sup,lwd=3,col='blue')

lm.mod<-sym.lm.bi(sym.var(lynne1,2),sym.var(lynne1,1),method='tops')
sym.scatterplot(sym.var(lynne1,2),sym.var(lynne1,1),labels=TRUE,
col='red',main='Linear Regression')
abline(lm.mod,lwd=3,col='blue')

lm.mod<-sym.lm.bi(sym.var(lynne1,2),sym.var(lynne1,1),method='billard')
sym.scatterplot(sym.var(lynne1,2),sym.var(lynne1,1),labels=TRUE,
col='red',main='Linear Regression')
abline(lm.mod$Intercept,lm.mod$Beta1,lwd=3,col='blue')