Plot lba objects

Description

S3 methods for lba objects. It's possible two types of visualisation: the lba type, suggested by van der Ark (1999) and correspondence analysis suggested by Jelihovschi (2011).

Usage

## S3 method for class 'lba'
plot(x,
     budget.prop = TRUE,
     col.points  = NULL,
     col.lines   = NULL,# only to K = 2
     col.budget  = NULL,
     pch.points  = NULL,
     pch.budget  = NULL,
     lty.lines   = NULL,# only to K = 2
     lty.budget  = NULL,
     lwd.lines   = NULL,# only to K = 2
     lwd.budget  = NULL,
     legend      = TRUE,
     with.ml     = c('mix','lat'),
     type        =c("lba","corr"),
     ...) 

Arguments

Details

Plot.lba plots two types of graphics. The first, (type = 'lba'), is the one suggested at de Leeuw et all (1990) and at van der Ark (1999) thesis. In this type, it is only possible the graphical views for K = 2 and K = 3. When K = 2, the heads of the latent budgets are be connected by a (one dimensional) line segment. When K = 3, the heads of the latent budgets are the vertices of a triangle, and the plot is made with help of triax.plot function of plotrix package. In the second type, plot.ca, suggested at Jelihovschi et all (2011), the graphical display is made by making use of correspondence analysis graphics of the mixing parameters and latent components matrices. This is done with the function ca of ca package. In this case, a graphic display is possible for K >= 3.

Author(s)

Enio G. Jelihovschi (eniojelihovs@gmail.com)
Ivan Bezerra Allaman (ivanalaman@gmail.com)

References

de Leeuw, J., van der Heijden, P.G.M., and Verboon, P. 1990. "A latent time budget model". Statistica Neerlandica. 44, 1, 1-21.

Jelihovschi, E.G., Alves, R.R., and Correa, F.M. 2011. Interacting latent budget analysis and correspondence analysis to analyze beauty salon management data. Biometric Brazilian Journal, 29, 657-673.

van der Ark, A. L. 1999. Contributions to Latent Budget Analysis, a tool for the analysis of comositional data. Ph.D. Thesis University of Utrecht.

See Also

triax.plot, triax.points, plot.ca.

Examples

data('votB')
ex1 <- lba(city ~ parties,
           data=votB,
           K = 2) 
plot(ex1)

# It's very simple. with colors!
plot(ex1,
     col.points=1:6,
     col.lines=1:6)

# I don't want your output. I want to provide my values and to use only your graphics. Ok!!!!
# Example: pag. 34, inner extreme solution
k1 <- c(.16,.68,.06,.45,.54,.30,.00,.45,1  ,.19,.62,.68,.57,.50,.30,
        .88,.46,.09,.62,.26,.53,.47,.58,.09,.26,.40,.70,.16,.71,.32)
k2 <- c(.84,.32,.94,.55,.46,.7 ,  1,.55,.00,.81,.38,.32,.43,.50,.7,
        .12,.54,.91,.38,.74,.47,.53,.42,.91,.74,.60,.30,.84,.29,.68)
my_mixing <- cbind(k1,k2)
rownames(my_mixing) <- paste(rep(c('H','C','R','Q','T','N','Se','Sk','L','V'),
                                 rep(3,10)),
                             c('+','=','-'),
                             sep = '')

my_budget <- matrix(c(.43,.57),
                    ncol=2)

colnames(my_budget) <- paste('LB',1:2,sep='')

my_data <- list(NA,
                NA,
                NA,
                NA,
                NA,
                my_mixing,
                NA,
                NA,
                my_budget)
class(my_data) = 'lba'

# I didn't like. I want equal of book. Pag. 41.

l_points <- c(2,24,17,
              rep(3,3),
              6,6,25,
              rep(4,3),
              1,16,21,
              rep(11,3),
              5,5,23,
              rep(8,3),
              rep(9,3),
              rep(13,3))

my_colors <- rep(c('red','blue','black'),
                 10)

plot(my_data,
     pch.points = l_points,
     col.points = my_colors,
     lty.budget = 2,
     pch.budget = 7,
     legend = FALSE)
legend(-0.1,
       1.2,
       rownames(my_mixing),
       pch = l_points,
       col = my_colors,
       xpd = TRUE,
       cex = 0.8,
       ncol = 10) # Is beautiful!! 

Results