Plot Concordance

A concordance plot of 2 numeric vectors, passed either as a 2-column data.frame/tibble object or separately as x and y.

For cc.type = "ccc", Lin's Concordance Correlation Coefficient as is used to estimate the correlation coefficient. For all other allowed values of cc.type, stats::cor.test() is used.

Usage

plotConcord(x, ...)

# S3 method for numeric
plotConcord(x, y, ...)

# S3 method for data.frame
plotConcord(
  x,
  x.lab = "x",
  y.lab = "y",
  main = NULL,
  pt.col = soma_purple,
  pt.size = 2,
  add.rug = TRUE,
  identify = FALSE,
  all.labels = seq_len(nrow(x)),
  spread = 1,
  label.size = 3,
  cc.type = c("ccc", "pearson", "spearman", "kendall"),
  ...
)

Arguments

x: Numeric. First vector comparison (x-axis). Alternatively, x can be a two column data frame containing vector data to compare.
...: For extensibility to S3 methods.
y: Numeric. Second vector comparison (y-axis). Omitted if x is a data.frame.
x.lab: Character. Optional string for the x-axis. Otherwise one is automatically generated (default).
y.lab: Character. Optional string for the y-axis. Otherwise one is automatically generated (default).
main: Character. Main title for the plot. See ggtitle() for ggplot2 style graphics.
pt.col: Character. The color to be used for the points.
pt.size: Numeric. The point size.
add.rug: Logical. If TRUE, axis includes ticks for each point.
identify: Logical. If TRUE, points are auto-identified.
all.labels: Character. The labels for each point if identify = TRUE. The points that are actually labeled is determined by the spread = argument. The default simply numbers the points.
spread: Numeric. The width from the unit line to start labeling points. The default (spread = 1) is reasonable for log10-transformed RFU data.
label.size: Numeric. The size for the point labels.
cc.type: Character. The correlation coefficient estimator to use. Must be one of c("ccc", "pearson", "spearman", "kendall").

Value

A concordance plot with the squared correlation coefficient and related statistics as a sub-title, a robust linear regression fit line, the unit line, and optionally labeled points.

References

Lin, Lawrence I-Kuei. 1989. A Concordance Correlation Coefficient to Evaluate Reproducibility. Biometrics. 45:255-268.

Author

Stu Field

Examples

x <- withr::with_seed(1234L, rnorm(500, 400, 30))

# bias
plotConcord(x, x + 20)


# scale
plotConcord(x, x * 1.2)


# random
y <- x + withr::with_seed(2345L, rnorm(500, sd = sd(x) / 2))
plotConcord(x, y - (y * 0.02))


# using Spearman's Correlation Coefficient
plotConcord(x, y - (y * 0.02), cc.type = "spearman")


# using Pearson's Correlation Coefficient
plotConcord(x, y - (y * 0.02), cc.type = "pearson")


# using Kendall's Correlation Coefficient
plotConcord(x, y - (y * 0.02), cc.type = "kendall")


# Label points by number (default)
plotConcord(x, y - (y * 0.02), identify = TRUE, spread = 25)
#> ✓ Auto-labeling 69 points by spread = 25 


# S3 data frame method
df <- data.frame(x = x, y = y - (y * 0.02))
plotConcord(df, identify = TRUE, spread = 25)
#> ✓ Auto-labeling 69 points by spread = 25