do_fada_multiv_robust: Run the whole archetypoid analysis with the functional multivariate robust Frobenius norm

Description

This function executes the entire procedure involved in the functional archetypoid analysis. Firstly, the initial vector of archetypoids is obtained using the functional archetypal algorithm and finally, the optimal vector of archetypoids is returned.

Usage

do_fada_multiv_robust(subset, numArchoid, numRep, huge, prob, compare = FALSE, PM,
                      method = "adjbox")

Arguments

subset

Data to obtain archetypes. In fadalara this is a subset of the entire data frame.

numArchoid

Number of archetypes/archetypoids.

numRep

For each numArch, run the archetype algorithm numRep times.

huge

Penalization to solve the convex least squares problem, see archetypoids.

prob

Probability with values in [0,1].

compare

Boolean argument to compute the non-robust residual sum of squares to compare these results with the ones provided by do_fada.

Penalty matrix obtained with eval.penalty.

method

Method to compute the outliers. So far the only option allowed is 'adjbox' for using adjusted boxplots for skewed distributions. The use of tolerance intervals might also be explored in the future for the multivariate case.

Value

A list with the following elements:

cases: Final vector of archetypoids.
alphas: Alpha coefficients for the final vector of archetypoids.
rss: Residual sum of squares corresponding to the final vector of archetypoids.
rss_non_rob: If compare=TRUE, this is the residual sum of squares using the non-robust Frobenius norm. Otherwise, NULL.
resid Vector of residuals.
outliers: Outliers.
local_rel_imp Matrix with the local (casewise) relative importance (in percentage) of each variable for the outlier identification. Only for the multivariate case. It is relative to the outlier observation itself. The other observations are not considered for computing this importance. This procedure works because the functional variables are in the same scale, after standardizing. Otherwise, it couldn't be interpreted like that.
margi_rel_imp Matrix with the marginal relative importance of each variable (in percentage) for the outlier identification. Only for the multivariate case. In this case, the other points are considered, since the value of the outlier observation is compared with the remaining points.

References

Moliner, J. and Epifanio, I., Robust multivariate and functional archetypal analysis with application to financial time series analysis, 2019. Physica A: Statistical Mechanics and its Applications 519, 195-208. https://doi.org/10.1016/j.physa.2018.12.036

Examples

Run this code

# NOT RUN {
library(fda)
?growth
str(growth)
hgtm <- growth$hgtm
hgtf <- growth$hgtf[,1:39]

# Create array:
nvars <- 2
data.array <- array(0, dim = c(dim(hgtm), nvars))
data.array[,,1] <- as.matrix(hgtm)
data.array[,,2] <- as.matrix(hgtf)
rownames(data.array) <- 1:nrow(hgtm)
colnames(data.array) <- colnames(hgtm)
str(data.array)

# Create basis:
nbasis <- 10
basis_fd <- create.bspline.basis(c(1,nrow(hgtm)), nbasis)
PM <- eval.penalty(basis_fd)
# Make fd object:
temp_points <- 1:nrow(hgtm)
temp_fd <- Data2fd(argvals = temp_points, y = data.array, basisobj = basis_fd)

X <- array(0, dim = c(dim(t(temp_fd$coefs[,,1])), nvars))
X[,,1] <- t(temp_fd$coef[,,1]) 
X[,,2] <- t(temp_fd$coef[,,2])

# Standardize the variables:
Xs <- X
Xs[,,1] <- scale(X[,,1])
Xs[,,2] <- scale(X[,,2])

suppressWarnings(RNGversion("3.5.0"))
set.seed(2018)
res_fada <- do_fada_multiv_robust(subset = Xs, numArchoid = 3, numRep = 5, huge = 200, 
                                  prob = 0.75, compare = FALSE, PM = PM, method = "adjbox")
str(res_fada)
res_fada$cases
#[1]  8 24 29
res_fada$rss
#[1] 2.301741
# }
# NOT RUN {
                                  
# }