plot_1d: Plotting Functions for AdaPT with 1D Covariates

Description Usage Arguments Examples

Description

Plotting the outputs of adapt when x is 1-dimensional, including threshold curves and level curves of local FDR.

Usage

1
2
3
4
5
plot_1d_thresh(obj, x, pvals, alpha, title, xlab = "x", xlim = NULL,
  disp_ymax = 0.2, num_yticks = 3, rand_seed_perturb = NA, ...)

plot_1d_lfdr(obj, x, pvals, alpha, title, xlab = "x", xlim = NULL,
  disp_ymax = 0.2, num_yticks = 3, legend_pos = "topright", ...)

Arguments

obj

an 'adapt' object

x

covariates (i.e. side-information). Should be compatible to models and 1-dimensional.

pvals

a vector of values in [0, 1]. P-values

alpha

a positive scalar in (0, 1). Target FDR level

title

a string. Title of the figure

xlab

a string. Label of the x-axis

xlim

a vector of length 2. Limits of x-axis

disp_ymax

a positive scalar in (0, 1]. Maximum value displayed in the y-axis

num_yticks

a positive integer. Number of ticks in the y-axis

rand_seed_perturb

random seed if jitter is added. NA if no jittering is needed

...

other arguments passed to par

legend_pos

a string. Position of the legend

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
# Load estrogen data
data(estrogen)
pvals <- as.numeric(estrogen$pvals)
x <- data.frame(x = as.numeric(estrogen$ord_high))
dist <- beta_family()

# Subsample the data for convenience
inds <- (x$x <= 5000)
pvals <- pvals[inds]
x <- x[inds,,drop = FALSE]

# Run adapt_glm
library("splines")
formulas <- paste0("ns(x, df = ", 6:10, ")")
res <- adapt_glm(x = x, pvals = pvals, pi_formulas = formulas,
                 mu_formulas = formulas, dist = dist, nfits = 10)

# Plots
par(mfrow = c(2, 1))
plot_1d_thresh(res, x, pvals, 0.1, "P-value Thresholds (alpha = 0.1)",
               disp_ymax = 0.5)
plot_1d_lfdr(res, x, pvals, 0.1, "Level Curves of lfdr (alpha = 0.1)",
             disp_ymax = 0.5)

adaptMT documentation built on May 1, 2019, 10:15 p.m.