Plot a trend on log-log paper.
LogLogPlot( frame, xvar, yvar, title, ..., use_coord_trans = FALSE, point_color = "black", linear_color = "#018571", quadratic_color = "#a6611a", smoothing_color = "blue" )
data frame to get values from
name of the independent (input or model) column in frame
name of the dependent (output or result to be modeled) column in frame
title to place on plot
no unnamed argument, added to force named binding of later arguments.
logical if TRUE, use coord_trans instead of
the color of the data points
the color of the linear growth lines
the color of the quadratic growth lines
the color of the smoothing line through the data
This plot is intended for plotting functions that are observed costs or durations as a function of problem size. In this case we expect the ideal or expected cost function to be non-decreasing. Any negative trends are assumed to arise from the noise model. The graph is specialized to compare non-decreasing linear and non-decreasing quadratic growth.
Some care must be taken in drawing conclusions from log-log plots, as the transform is fairly violent. Please see: "(Mar's Law) Everything is linear if plotted log-log with a fat magic marker" (from Akin's Laws of Spacecraft Design http://spacecraft.ssl.umd.edu/akins_laws.html), and "So You Think You Have a Power Law" http://bactra.org/weblog/491.html.
set.seed(5326) frm = data.frame(x = 1:20) frm$y <- 5 + frm$x + 0.2 * frm$x * frm$x + 0.1*abs(rnorm(nrow(frm))) WVPlots::LogLogPlot(frm, "x", "y", title="Example Trend")#>