A line plot showing the cumulative proportion of explained variance as a function of the number of Principal Component Analysis (PCA) components. This visualization helps determine the optimal number of components to retain by displaying the trade-off between dimensionality reduction and information preservation. The cumulative curve typically exhibits an elbow pattern where additional components yield diminishing returns, and horizontal threshold lines (e.g., 90%, 95%) guide component selection decisions.
#' anyplot.ai
#' line-pca-variance-cumulative: Cumulative Explained Variance for PCA Component Selection
#' Library: ggplot2 3.5.1 | R 4.4.1
#' Quality: 88/100 | Created: 2026-05-29
library(ggplot2)
library(ragg)
set.seed(42)
# --- Theme tokens -----------------------------------------------------------
THEME <- Sys.getenv("ANYPLOT_THEME", "light")
PAGE_BG <- if (THEME == "light") "#FAF8F1" else "#1A1A17"
ELEVATED_BG <- if (THEME == "light") "#FFFDF6" else "#242420"
INK <- if (THEME == "light") "#1A1A17" else "#F0EFE8"
INK_SOFT <- if (THEME == "light") "#4A4A44" else "#B8B7B0"
INK_MUTED <- if (THEME == "light") "#6B6A63" else "#A8A79F"
IMPRINT_PALETTE <- c(
"#009E73", "#C475FD", "#4467A3", "#BD8233",
"#AE3030", "#2ABCCD", "#954477", "#99B314"
)
# --- Data -------------------------------------------------------------------
# 300-sample × 15-feature sensor dataset driven by 3 latent factors,
# giving a realistic PCA scree with a clear elbow around components 5-7
n_obs <- 300
n_features <- 15
f1 <- rnorm(n_obs)
f2 <- rnorm(n_obs)
f3 <- rnorm(n_obs)
X <- matrix(0, nrow = n_obs, ncol = n_features)
for (i in 1:5) X[, i] <- f1 * (1.8 - 0.25 * i) + rnorm(n_obs, 0, 0.3)
for (i in 6:10) X[, i] <- f2 * (1.3 - 0.12 * (i - 5)) + rnorm(n_obs, 0, 0.4)
for (i in 11:15) X[, i] <- f3 * (0.9 - 0.08 * (i - 10)) + rnorm(n_obs, 0, 0.5)
pca <- prcomp(X, center = TRUE, scale. = TRUE)
var_ind <- pca$sdev^2 / sum(pca$sdev^2) * 100
var_cum <- cumsum(var_ind)
df <- data.frame(
component = seq_len(n_features),
individual_pct = var_ind,
cumulative_pct = var_cum
)
thr90 <- min(which(var_cum >= 90))
thr95 <- min(which(var_cum >= 95))
# Elbow detection: point of maximum perpendicular distance from line
# connecting first and last cumulative-variance values
a_coef <- var_cum[n_features] - var_cum[1]
b_coef <- -(n_features - 1)
c_coef <- (n_features - 1) * var_cum[1] - (var_cum[n_features] - var_cum[1]) * 1
knee_dist <- abs(a_coef * seq_len(n_features) + b_coef * var_cum + c_coef) /
sqrt(a_coef^2 + b_coef^2)
elbow_pc <- which.max(knee_dist)
# --- Plot -------------------------------------------------------------------
title_str <- "line-pca-variance-cumulative · r · ggplot2 · anyplot.ai"
p <- ggplot(df, aes(x = component)) +
geom_col(aes(y = individual_pct),
fill = IMPRINT_PALETTE[1], alpha = 0.13, width = 0.65) +
geom_hline(yintercept = 90, linetype = "dashed",
color = INK_SOFT, linewidth = 0.55) +
geom_hline(yintercept = 95, linetype = "dashed",
color = INK_SOFT, linewidth = 0.55) +
geom_vline(xintercept = elbow_pc, linetype = "dotted",
color = IMPRINT_PALETTE[1], linewidth = 0.65, alpha = 0.55) +
geom_line(aes(y = cumulative_pct),
color = IMPRINT_PALETTE[1], linewidth = 1.2) +
geom_point(aes(y = cumulative_pct),
color = IMPRINT_PALETTE[1], size = 2.8) +
annotate("text", x = 1.2, y = 91.8,
label = "90%", color = INK_MUTED, size = 3.3, hjust = 0) +
annotate("text", x = 1.2, y = 96.8,
label = "95%", color = INK_MUTED, size = 3.3, hjust = 0) +
annotate("text", x = thr90 + 0.3, y = 85,
label = sprintf("PC%d\n90%%", thr90),
color = INK_MUTED, size = 3.0, hjust = 0, lineheight = 0.9) +
annotate("text", x = elbow_pc + 0.35, y = 5,
label = sprintf("PC%d\nelbow", elbow_pc),
color = IMPRINT_PALETTE[1], size = 3.3, hjust = 0, lineheight = 0.9) +
scale_x_continuous(breaks = seq_len(n_features)) +
scale_y_continuous(
limits = c(0, 105),
breaks = seq(0, 100, 20),
labels = function(x) paste0(x, "%")
) +
labs(
x = "Number of Principal Components",
y = "Cumulative Explained Variance",
title = title_str
) +
theme_minimal(base_size = 8) +
theme(
plot.background = element_rect(fill = PAGE_BG, color = PAGE_BG),
panel.background = element_rect(fill = PAGE_BG, color = NA),
panel.grid.major.y = element_line(color = INK, linewidth = 0.18),
panel.grid.major.x = element_blank(),
panel.grid.minor = element_blank(),
panel.border = element_blank(),
axis.line = element_line(color = INK_SOFT, linewidth = 0.45),
axis.ticks = element_blank(),
axis.title = element_text(color = INK, size = 10),
axis.text = element_text(color = INK_SOFT, size = 8),
plot.title = element_text(color = INK, size = 12, face = "bold",
margin = margin(b = 14)),
plot.margin = margin(22, 24, 18, 18)
)
# --- Save -------------------------------------------------------------------
ggsave(
filename = sprintf("plot-%s.png", THEME),
plot = p,
device = ragg::agg_png,
width = 8,
height = 4.5,
units = "in",
dpi = 400
)
Runnable source as JSON, for any HTTP client: https://api.anyplot.ai/specs/line-pca-variance-cumulative/ggplot2/code. Any spec id and library id listed in llms-full.txt fit the same URL shape; every URL below is complete and callable.
{
"spec_id": "line-pca-variance-cumulative",
"language": "r",
"library": "ggplot2",
"page": "https://anyplot.ai/line-pca-variance-cumulative/r/ggplot2",
"hub": "https://anyplot.ai/line-pca-variance-cumulative",
"code_json": "https://api.anyplot.ai/specs/line-pca-variance-cumulative/ggplot2/code",
"spec_json": "https://api.anyplot.ai/specs/line-pca-variance-cumulative",
"render_light_png": "https://storage.googleapis.com/anyplot-images/plots/line-pca-variance-cumulative/r/ggplot2/plot-light.png",
"render_dark_png": "https://storage.googleapis.com/anyplot-images/plots/line-pca-variance-cumulative/r/ggplot2/plot-dark.png",
"quality_score": 88.0,
"license": "MIT",
"guide": "https://anyplot.ai/llms.txt"
}Part of Cumulative Explained Variance for PCA Component Selection on anyplot.ai.