Minor updates

R/ellipseCoord.R

@@ -27,8 +27,6 @@
 #' xy_coord <- ellipseCoord(data = pca_scores, pcx = 1, pcy = 2, conf.limit = 0.95, pts = 200)
 #'
 ellipseCoord <- function(data, pcx = 1, pcy = 2, conf.limit = 0.95, pts = 200) {
-
-  # check input validity
   if (length(data) == 0) {
     stop("Data must not be empty.")
   }
@@ -44,32 +42,20 @@ ellipseCoord <- function(data, pcx = 1, pcy = 2, conf.limit = 0.95, pts = 200) {
   if (conf.limit < 0 || conf.limit > 1) {
     stop("Confidence level should be between 0 and 1.")
   }
-
-  # matrix of data
   X <- as.matrix(data)
-
-  # Sample size
   n <- nrow(X)
-
-  # Confidence limit
   alpha <- as.numeric(conf.limit)
-
-  # Number of points
   m <- as.numeric(pts)
   p <- seq(0, 2*pi, length.out = m)
 
-  # # Hotelling’s T-square limit
   Tsq_limit <- (2*(n-1)/(n-2))*stats::qf(p = alpha, df1 = 2, df2 = (n-2))
 
-  # Coordinate points
   rx <- sqrt(Tsq_limit*stats::var(X[, pcx]))
   ry <- sqrt(Tsq_limit*stats::var(X[, pcy]))
 
   res.coord <- tibble::tibble(
     x = rx*cos(p) + mean(X[, pcx], na.rm = TRUE),
     y = ry*sin(p) + mean(X[, pcy], na.rm = TRUE)
   )
-
   return(res.coord)
-
 }

R/ellipseParam.R

@@ -12,7 +12,6 @@
 #'4. **cutoff.95pct** Number corresponding to the T-squared cutoff at 95% confidence level.
 #' @export ellipseParam
 #' @author Christian L. Goueguel,
-#' christian.goueguel@gmail.com
 #' @examples
 #' ## Principal components analysis (PCA)
 #' library(dplyr)
@@ -31,8 +30,6 @@
 #' T2 <- ellipseParam(data = pca_scores, k = 2, pcx = 1, pcy = 2)
 #'
 ellipseParam <- function(data, k = 2, pcx = 1, pcy = 2) {
-
-  # check input validity
   if (length(data) == 0) {
     stop("Data must not be empty.")
   }
@@ -45,52 +42,37 @@ ellipseParam <- function(data, k = 2, pcx = 1, pcy = 2) {
   if (pcx == 0 || pcy == 0) {
     stop("pcx and pcy must be non-zero.")
   }
-  if (k < 2) {
-    stop("k must be at least equal to 2.")
+  if (!is.numeric(k) || k < 2 || k != floor(k)) {
+    stop("k must be a positive integer >= 2.")
   }
   if (k > ncol(data)) {
-    stop("k exceeds the number of component in the data.")
+    stop("k exceeds the number of columns in the data.")
   }
-
-  # matrix of data
   X <- as.matrix(data)
-
-  # sample size
   n <- nrow(X)
-
-  # number of principal component
   A <- as.numeric(k)
-
-  # Squared Mahalanobis distance
   MDsq <- stats::mahalanobis(
     x = X,
     center = colMeans(X),
     cov = stats::cov(X),
     inverted = FALSE
   )
-
   if(k > 2) {
-    # 99% and 95% confidence limit
     Tsq_limit1 <- (A*(n-1)/(n-A))*stats::qf(p = 0.99, df1 = A, df2 = (n-A))
     Tsq_limit2 <- (A*(n-1)/(n-A))*stats::qf(p = 0.95, df1 = A, df2 = (n-A))
-
-    # Hotelling’s T-square
     Tsq <- tibble::tibble(value = ((n-A)/(A*(n-1)))*MDsq)
 
     res_list <- list(
       "Tsquare" = Tsq,
       "cutoff.99pct" = Tsq_limit1,
       "cutoff.95pct" = Tsq_limit2
     )
+
     return(res_list)
     }
-
   if(k == 2) {
-    # 99% and 95% confidence limit
     Tsq_limit1 <- (A*(n-1)/(n-A))*stats::qf(p = 0.99, df1 = A, df2 = (n-A))
     Tsq_limit2 <- (A*(n-1)/(n-A))*stats::qf(p = 0.95, df1 = A, df2 = (n-A))
-
-    # Hotelling ellipse semi-axes
     a_limit1 <- sqrt(Tsq_limit1*stats::var(X[, pcx]))
     a_limit2 <- sqrt(Tsq_limit2*stats::var(X[, pcx]))
     b_limit1 <- sqrt(Tsq_limit1*stats::var(X[, pcy]))
@@ -103,7 +85,6 @@ ellipseParam <- function(data, k = 2, pcx = 1, pcy = 2) {
       b.95pct = b_limit2
     )
 
-    # Hotelling’s T-square
     Tsq <- tibble::tibble(value = ((n-2)/(2*(n-1)))*MDsq)
 
     res_list <- list(
@@ -112,7 +93,7 @@ ellipseParam <- function(data, k = 2, pcx = 1, pcy = 2) {
       "cutoff.99pct" = Tsq_limit1,
       "cutoff.95pct" = Tsq_limit2
     )
+
     return(res_list)
     }
   }
-