Figure6_Plot

Scenario 11

Scenario 11 (Varying \(p\)): \(n = \textrm{500}\), \(p \in [\textrm{20},\textrm{20,000}]\), \(s = \textrm{20}\), \(\textrm{Design} = \textrm{IndepGauss}\), \(\textrm{PVE} = 0.5\) and \(\textrm{SignalShape} = \textrm{PointNormal}\).

res_df       = readRDS("results/scenario11.RDS")
method_list  = c("Mr.ASH","varbvs","BayesB","B-Lasso","SuSiE","E-NET","Lasso",
                 "Ridge","SCAD.2tuning","MCP.2tuning","SCAD.1tuning","MCP.1tuning","L0Learn")
df           = data.frame()
p_range      = c(20,50,100,200,500,1000,2000,5000,10000,20000)
for (i in 1:10) {
res_df[[i]]$fit   = rep(method_list, each = 20)
res_df[[i]]$fit   = factor(res_df[[i]]$fit, levels = method_list)
df                = rbind(df, data.frame(pred = c(colMeans(matrix(res_df[[i]]$pred, 20, 13))),
                                         time2 = apply(matrix(res_df[[i]]$time, 20, 13), 2, median),
                                         time = c(colMeans(matrix(res_df[[i]]$time, 20, 13))),
                                         p    = p_range[i],
                                         fit  = method_list))
}
df$fit = factor(df$fit, levels = method_list)
df = df[df$fit %in% c("Mr.ASH","E-NET","Lasso","Ridge","SCAD.2tuning",
                      "MCP.2tuning","SCAD.1tuning","MCP.1tuning","L0Learn"),]
df = df[df$p %in% c(20,50,200,500,1000,2000,10000,20000),]
df$size = 0.5
df$size[1:8] = 1.2
col   = gg_color_hue(13)[c(1:6,5,6,7)]
df$lt = "solid"
df$lt[49:64] = "dashed"
shape = c(19,17,24,25,9,3,11,4,5,7,8,10,1)[c(1:6,5,6,7)]
p1    = ggplot(df) + geom_line(aes(x = p, y = pred, color = fit), size = df$size, linetype = df$lt) +
  geom_point(aes(x = p, y = pred, color = fit, shape = fit), size = 2.5) +
  theme_cowplot(font_size = 22) +
  scale_x_continuous(trans = "log10", breaks = c(20,50,200,500,1000,2000,10000,20000)) +
  labs(y = "predictior error (rmse / sigma)", x = "number of coefficients (p)") +
  theme(axis.line = element_blank(),
        plot.title = element_text(hjust = 0.5),
        axis.title.x = element_blank(),
        axis.text.x  = element_text(angle = 45,hjust = 1),
        axis.title.y = element_blank()) +
  scale_color_manual(values = col) +
  scale_shape_manual(values = shape) +
  scale_y_continuous(trans = "log10", breaks = c(1,1.1,1.2,1.3,1.4)) +
  coord_cartesian(ylim = c(1,1.3))
fig_dummy4 = p1
fig_main = p1 + theme(legend.position = "none")
suptitle   = ggdraw() + draw_label("Varying p", fontface  = 'bold', size = 14) 
subtitle   = ggdraw() + draw_label("Scenario: IndepGauss + PointNormal, n = 500, p = 20-20000, s = 20, pve = 0.5", fontface  = 'bold', size = 14) 
fig11_1    = plot_grid(fig_main, ncol = 1, rel_heights = c(0.05,0.95))

res_df       = readRDS("results/scenario11.RDS")
method_list  = c("Mr.ASH","varbvs","BayesB","B-Lasso","SuSiE","E-NET","Lasso",
                 "Ridge","SCAD.2tuning","MCP.2tuning","SCAD.1tuning","MCP.1tuning","L0Learn")
df           = data.frame()
p_range      = c(20,50,100,200,500,1000,2000,5000,10000,20000)
for (i in 1:10) {
res_df[[i]]$fit   = rep(method_list, each = 20)
res_df[[i]]$fit   = factor(res_df[[i]]$fit, levels = method_list)
df                = rbind(df, data.frame(pred = c(colMeans(matrix(res_df[[i]]$pred, 20, 13))),
                                         time2 = apply(matrix(res_df[[i]]$time, 20, 13), 2, median),
                                         time = c(colMeans(matrix(res_df[[i]]$time, 20, 13))),
                                         p    = p_range[i],
                                         fit  = method_list))
}
df$fit = factor(df$fit, levels = method_list)
df = df[df$fit %in% c("Mr.ASH","E-NET","Lasso","Ridge","SCAD.2tuning",
                      "MCP.2tuning","SCAD.1tuning","MCP.1tuning","L0Learn"),]
df$size = 0.5
df$size[1:5] = 1.2; df$size[46:50] = 1.2;
col   = gg_color_hue(13)[c(1:6,5,6,7)]
df$lt = "solid"
df$lt[c(31:40,76:85)] = "dashed"
shape = c(19,17,24,25,9,3,11,4,5,7,8,1,2,6)[c(1:6,5,6,7)]
p1    = ggplot() + geom_line(aes(x = df$p[df$p <= 500], y = df$time[df$p <= 500], color = df$fit[df$p <= 500]),
                                 size = df$size[df$p <= 500], linetype = df$lt[df$p <= 500]) +
  geom_point(aes(x = df$p[df$p <= 500], y = df$time[df$p <= 500], color = df$fit[df$p <= 500],
                 shape = df$fit[df$p <= 500]), size = 2.5) +
  scale_color_manual(values = col) +
  scale_shape_manual(values = shape)
fig_main = p1 + geom_line(aes(x = df$p[df$p > 500], y = df$time[df$p > 500], color = df$fit[df$p > 500]),
                              size = df$size[df$p > 500], linetype = df$lt[df$p > 500]) + 
    geom_point(aes(x = df$p[df$p > 500], y = df$time[df$p > 500], color = df$fit[df$p > 500],
                   shape = df$fit[df$p > 500]), size = 2.5) +
  theme_cowplot(font_size = 22) +
  scale_x_continuous(trans = "log10", breaks = p_range) +
  labs(y = "computation time (seconds)", x = "number of nonzero coefficients (s)") +
  theme(axis.line = element_blank(),
        plot.title = element_text(hjust = 0.5),
        legend.position = "none",
        axis.title.x = element_blank(),
        axis.text.x  = element_text(angle = 45,hjust = 1),
        axis.title.y = element_blank()) +
  scale_y_continuous(trans = "log10", breaks = c(0.01,0.1,1,10,100)) +
  coord_cartesian(ylim = c(0.01,300))
subtitle   = ggdraw() + draw_label("Computation Time (seconds)", fontface = 'bold', size = 14) 
fig11_2    = plot_grid(fig_main, ncol = 1, rel_heights = c(0.05,0.95))

res_df       = readRDS("./results/scenario11.RDS")
method_list  = c("Mr.ASH","varbvs","BayesB","B-Lasso","SuSiE","E-NET","Lasso",
                 "Ridge","SCAD.2tuning","MCP.2tuning","SCAD.1tuning","MCP.1tuning","L0Learn")
df           = data.frame()
p_range      = c(20,50,100,200,500,1000,2000,5000,10000,20000)
for (i in 1:10) {
res_df[[i]]$fit   = rep(method_list, each = 20)
res_df[[i]]$fit   = factor(res_df[[i]]$fit, levels = method_list)
df                = rbind(df, data.frame(pred = c(colMeans(matrix(res_df[[i]]$pred, 20, 13))),
                                         time2 = apply(matrix(res_df[[i]]$time, 20, 13), 2, median),
                                         time = c(colMeans(matrix(res_df[[i]]$time, 20, 13))),
                                         p    = p_range[i],
                                         fit  = method_list))
}
df$fit = factor(df$fit, levels = c("Mr.ASH","E-NET","Lasso", "Ridge","SCAD.2tuning","MCP.2tuning","L0Learn",
                                   "varbvs","BayesB","B-Lasso","SuSiE","SCAD.1tuning","MCP.1tuning"))
df = df[df$p %in% c(20,50,200,500,1000,2000,10000,20000),]
df$size = 0.5
df$size[1:8] = 1.2
col   = gg_color_hue(13)[c(1:11,5,6)]
df$lt = "solid"
shape = c(19,17,24,25,9,3,11,4,5,7,8,10,1)[c(1:11,5,6)]
df$lt[89:104] = "dashed"
df$Method = df$fit
fig_dummy4 = ggplot(df) + geom_line(aes(x = p, y = pred, color = Method), size = df$size, linetype = df$lt) +
  geom_point(aes(x = p, y = pred, color = Method, shape = Method), size = 2.5) +
  theme_cowplot(font_size = 22) +
  scale_x_continuous(trans = "log10", breaks = c(20,50,200,500,1000,2000,10000,20000)) +
  labs(y = "predictior error (rmse / sigma)", x = "number of coefficients (p)") +
  theme(axis.line = element_blank(),
        plot.title = element_text(hjust = 0.5),
        axis.title.x = element_blank(),
        axis.text.x  = element_text(angle = 45,hjust = 1),
        axis.title.y = element_blank()) +
  scale_color_manual(values = col) +
  scale_shape_manual(values = shape) +
  scale_y_continuous(trans = "log10", breaks = c(1,1.1,1.2,1.3,1.4)) +
  coord_cartesian(ylim = c(1,1.3))

res_df       = readRDS("results/scenario11.RDS")
method_list  = c("Mr.ASH","varbvs","BayesB","B-Lasso","SuSiE","E-NET","Lasso",
                 "Ridge","SCAD.2tuning","MCP.2tuning","SCAD.1tuning","MCP.1tuning","L0Learn")
df           = data.frame()
p_range      = c(20,50,100,200,500,1000,2000,5000,10000,20000)
for (i in 1:10) {
res_df[[i]]$fit   = rep(method_list, each = 20)
res_df[[i]]$fit   = factor(res_df[[i]]$fit, levels = method_list)
df                = rbind(df, data.frame(pred = c(colMeans(matrix(res_df[[i]]$pred, 20, 13))),
                                         time2 = apply(matrix(res_df[[i]]$time, 20, 13), 2, median),
                                         time = c(colMeans(matrix(res_df[[i]]$time, 20, 13))),
                                         p    = p_range[i],
                                         fit  = method_list))
}
df$fit = factor(df$fit, levels = method_list)
df = df[df$p %in% c(20,50,200,500,1000,2000,10000,20000),]
df = df[df$fit %in% c("Mr.ASH","varbvs","BayesB","B-Lasso","SuSiE"),]
df$size = 0.5
df$size[1:8] = 1.2
col   = gg_color_hue(13)[c(1,8:11)]
df$lt = "solid"
shape = c(19,17,24,25,9,3,11,4,5,7,8,10,1)[c(1,8:11)]
p1    = ggplot(df) + geom_line(aes(x = p, y = pred, color = fit), size = df$size, linetype = df$lt) +
  geom_point(aes(x = p, y = pred, color = fit, shape = fit), size = 2.5) +
  theme_cowplot(font_size = 22) +
  scale_x_continuous(trans = "log10", breaks = c(20,50,200,500,1000,2000,10000,20000)) +
  labs(y = "predictior error (rmse / sigma)", x = "number of coefficients (p)") +
  theme(axis.line = element_blank(),
        plot.title = element_text(hjust = 0.5),
        axis.title.x = element_blank(),
        axis.text.x  = element_text(angle = 45,hjust = 1),
        axis.title.y = element_blank()) +
  scale_color_manual(values = col) +
  scale_shape_manual(values = shape) +
  scale_y_continuous(trans = "log10", breaks = c(1,1.1,1.2,1.3,1.4)) +
  coord_cartesian(ylim = c(1,1.3))
fig_main = p1 + theme(legend.position = "none")
# suptitle   = ggdraw() + draw_label("Varying p", fontface  = 'bold', size = 14) 
# subtitle   = ggdraw() + draw_label("Scenario: IndepGauss + PointNormal, n = 500, p = 20-20000, s = 20, pve = 0.5", fontface  = 'bold', size = 14) 
fig11_3    = plot_grid(fig_main, ncol = 1, rel_heights = c(0.05,0.95))

res_df       = readRDS("results/scenario11.RDS")
method_list  = c("Mr.ASH","varbvs","BayesB","B-Lasso","SuSiE","E-NET","Lasso",
                 "Ridge","SCAD.2tuning","MCP.2tuning","SCAD.1tuning","MCP.1tuning","L0Learn")
df           = data.frame()
p_range      = c(20,50,100,200,500,1000,2000,5000,10000,20000)
for (i in 1:10) {
res_df[[i]]$fit   = rep(method_list, each = 20)
res_df[[i]]$fit   = factor(res_df[[i]]$fit, levels = method_list)
df                = rbind(df, data.frame(pred = c(colMeans(matrix(res_df[[i]]$pred, 20, 13))),
                                         time2 = apply(matrix(res_df[[i]]$time, 20, 13), 2, median),
                                         time = c(colMeans(matrix(res_df[[i]]$time, 20, 13))),
                                         p    = p_range[i],
                                         fit  = method_list))
}
df$fit = factor(df$fit, levels = c("Mr.ASH","E-NET","Lasso", "Ridge","SCAD.1tuning","MCP.1tuning","L0Learn",
                      "varbvs","BayesB","B-Lasso","SuSiE","SCAD.2tuning","MCP.2tuning"))
df = df[df$fit %in% c("Mr.ASH","varbvs","BayesB","B-Lasso","SuSiE"),]
df$size = 0.5
df$size[1:9] = 1.2
col   = gg_color_hue(13)[c(1,8:11)]
shape = c(19,17,24,25,9,3,11,4,5,7,8,10,1)[c(1,8:11)]
fig_main = ggplot(df) + geom_line(aes(x = p, y = time, color = fit),
                              size = df$size) + 
    geom_point(aes(x = p, y = time, color = fit, shape = fit), size = 2.5) +
  theme_cowplot(font_size = 22) +
  scale_x_continuous(trans = "log10", breaks = p_range) +
  labs(y = "computation time (seconds)", x = "number of nonzero coefficients (s)") +
  theme(axis.line = element_blank(),
        plot.title = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x  = element_text(angle = 45,hjust = 1),
        axis.title.x = element_blank(),
        axis.title.y = element_blank()) +
  scale_y_continuous(trans = "log10", breaks = c(0.01,0.1,1,10,100), limits = c(0.01,300)) +
  scale_color_manual(values = col) +
  scale_shape_manual(values = shape)
#subtitle   = ggdraw() + draw_label("Computation Time (seconds)", fontface = 'bold', size = 14) 
fig11_4    = plot_grid(fig_main, ncol = 1, rel_heights = c(0.05,0.95))