plot.snowboot {snowboot}

Plot Degree Distribution Estimates

Description

Plot LSMI-based point estimates of probabilities of node degrees, \hat{f}(k), and of mean degree, \hat{\mu}, where k = 0, 1, \ldots are the degrees. The point estimates are supplemented with box-and-whisker plots of bootstrapped values (if the input is a boot_dd output) or element-wise bootstrap confidence intervals (if the input is a boot_ci output). See Chen et al. (2018).

Usage

## S3 method for class 'snowboot'
plot(
  x,
  k = NULL,
  plotmu = TRUE,
  plotlegend = TRUE,
  col0 = "gray50",
  lwd0 = 1,
  colpt = "royalblue3",
  lwdpt = 2,
  pchpt = 4,
  coli = "palegreen3",
  colibg = "palegreen",
  length = 0.1,
  boxwex = 0.4,
  legendargs = list(x = "topright", cex = 0.9, bty = "n"),
  las = 1,
  ...
)

Arguments

x

output of lsmi_dd, boot_dd, or boot_ci.

k

an integer vector with degrees to plot. By default, all degrees represented in x are plotted.

plotmu

logical value indicating whether to plot the results for mean degree (default is TRUE).

plotlegend

logical value indicating whether to plot a legend (default is TRUE).

col0

color to plot horizontal zero-line at f(k) = 0. Use NA for no plotting.

lwd0

width of the horizontal zero-line at f(k) = 0.

colpt

color for plotting point estimates.

lwdpt

line width for plotting point estimates.

pchpt

point type for plotting point estimates (see argument pch in points).

coli

color for plotting lines or borders of box-plots for bootstrap estimates.

colibg

background color, if plotting boxplots of bootstrapped estimates (see argument border in boxplot).

length

length of arrows, if plotting bootstrap confidence intervals (see argument length in arrows).

boxwex

argument of boxplot function.

legendargs

additional arguments for plotting the legend (see arguments in legend).

las

argument of plot function.

...

additional arguments to pass to the plot function.

References

Chen Y, Gel YR, Lyubchich V, Nezafati K (2018). “Snowboot: bootstrap methods for network inference.” The R Journal, 10(2), 95–113. doi: 10.32614/RJ-2018-056.

Examples

net <- artificial_networks[[1]]
x <- lsmi_dd(net = net, n.wave = 2, n.seed = 40)
plot(x)

x2 <- boot_dd(x)
plot(x2, k = c(1:10))

x3 <- boot_ci(x2, prob = 0.99)
plot(x3, k = c(1:10))