1d:T7bd,# Cleaning multivariate functional outliers Cleaning of the most remarkable multivariate functional outliers. This improves the performance of the archetypoid algorithm since it is not affected by spurious points. ```r do_clean_multiv(data, num_pts, range = 1.5, out_perc = 80, nbasis, nvars) ``` ## Arguments - `data`: Data frame with (temporal) points in the rows and observations in the columns. - `num_pts`: Number of temporal points. - `range`: Same parameter as in function `boxplot`. A value of 1.5 is enough to detect amplitude and shift outliers, while a value of 3 is needed to detect isolated outliers. - `out_perc`: Minimum number of temporal points (in percentage) to consider the observation as an outlier. Needed when `range=1.5`. - `nbasis`: Number of basis. - `nvars`: Number of variables. ## Returns List with the outliers for each variable. ## Examples ```r ## 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]) x1 <- t(Xs[,,1]) for (i in 2:nvars) { x12 <- t(Xs[,,i]) x1 <- rbind(x1, x12) } data_all <- t(x1) num_pts <- ncol(data_all) / nvars range <- 3 outl <- do_clean_multiv(t(data_all), num_pts, range, out_perc, nbasis, nvars) outl ## End(Not run) ``` ## See Also `boxplot` ## Author(s) Irene Epifanio

do_clean_multiv function