plspm (version 0.4.9)

it.reb: Iterative steps of Response-Based Unit Segmentation (REBUS)

Description

REBUS-PLS is an iterative algorithm for performing response based clustering in a PLS-PM framework. it.reb allows to perform the iterative steps of the REBUS-PLS Algorithm. It provides summarized results for final local models and the final partition of the units. Before running this function, it is necessary to run the res.clus function to choose the number of classes to take into account.

Usage

it.reb(pls, hclus.res, nk, Y = NULL, stop.crit = 0.005,
    iter.max = 100)

Arguments

pls

an object of class "plspm"

hclus.res

object of class "res.clus" returned by res.clus

nk

integer larger than 1 indicating the number of classes. This value should be defined according to the dendrogram obtained by performing res.clus.

Y

optional data matrix used when pls$data is NULL

stop.crit

Number indicating the stop criterion for the iterative algorithm. It is suggested to use the threshold of less than 0.05% of units changing class from one iteration to the other as stopping rule.

iter.max

integer indicating the maximum number of iterations

Value

an object of class "rebus"

loadings

Matrix of standardized loadings (i.e. correlations with LVs.) for each local model

path.coefs

Matrix of path coefficients for each local model

quality

Matrix containing the average communalities, the average redundancies, the R2 values, and the GoF index for each local model

segments

Vector defining the class membership of each unit

origdata.clas

The numeric matrix with original data and with a new column defining class membership of each unit

References

Esposito Vinzi, V., Trinchera, L., Squillacciotti, S., and Tenenhaus, M. (2008) REBUS-PLS: A Response-Based Procedure for detecting Unit Segments in PLS Path Modeling. Applied Stochastic Models in Business and Industry (ASMBI), 24, pp. 439-458. Trinchera, L. (2007) Unobserved Heterogeneity in Structural Equation Models: a new approach to latent class detection in PLS Path Modeling. Ph.D. Thesis, University of Naples "Federico II", Naples, Italy. http://www.fedoa.unina.it/2702/1/Trinchera_Statistica.pdf

See Also

plspm, rebus.pls, res.clus

Examples

## Not run: ------------------------------------
# ## Example of REBUS PLS with simulated data
# 
# # load simdata
# data("simdata", package='plspm')
# 
# # Calculate global plspm
# sim_inner = matrix(c(0,0,0,0,0,0,1,1,0), 3, 3, byrow=TRUE)
# dimnames(sim_inner) = list(c("Price", "Quality", "Satisfaction"),
#                            c("Price", "Quality", "Satisfaction"))
# sim_outer = list(c(1,2,3,4,5), c(6,7,8,9,10), c(11,12,13))
# sim_mod = c("A", "A", "A")  # reflective indicators
# sim_global = plspm(simdata, sim_inner,
#                    sim_outer, modes=sim_mod)
# sim_global
# 
# ## Then compute cluster analysis on residuals of global model
# sim_clus = res.clus(sim_global)
# 
# ## To complete REBUS, run iterative algorithm
# rebus_sim = it.reb(sim_global, sim_clus, nk=2,
#                    stop.crit=0.005, iter.max=100)
# 
# ## You can also compute complete outputs
# ## for local models by running:
# local_rebus = local.models(sim_global, rebus_sim)
# 
# # Display plspm summary for first local model
# summary(local_rebus$loc.model.1)
## ---------------------------------------------