/
plm_methods.R
668 lines (640 loc) · 37.3 KB
/
plm_methods.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
print_fun <- function(x){
cat(paste0(class(x)," - Call:\n"),
paste(deparse(x$formula), collapse = "\n"), "\n")
}
summary_fun <- function(x){
param_summary <- x$param_summary[,c('lower','median','upper')]
names(param_summary) <- paste0(names(param_summary),c('-2.5%','-50%','-97.5%'))
cat("\nFormula: \n",
paste(deparse(x$formula), sep = "\n", collapse = "\n"))
cat("\nLatent parameters:\n")
print(param_summary[1:2,],row.names = TRUE,digits=3,right=FALSE)
cat("\nHyperparameters:\n")
print(param_summary[3:nrow(param_summary),],row.names = TRUE,digits=3,right=FALSE)
cat("\nWAIC:",x$WAIC)
}
#' Custom bdrc theme
#'
#' @param ... not used in this function
#' @param scaling a numerical value which can be used to scale up or down the size of the text and titles of a plot that uses \code{theme_bdrc}. Defaults to 1.
#' @return returns a theme object for the package
#' @importFrom ggplot2 %+replace% theme_classic theme element_text element_blank element_rect
#' @keywords internal
theme_bdrc <- function(...,scaling=1){
title_size <- scaling*12
text_size <- scaling*12
plot_title_size=scaling*12
theme_classic() %+replace%
theme( #text = element_text(family="Times", face="plain"),
strip.background = element_blank(),
strip.text.x = element_text(size = title_size),
axis.title.x = element_text(size = title_size),
axis.title.y = element_text(size = title_size,angle=90),
axis.text.x = element_text(size = text_size),
axis.text.y = element_text(size = text_size),
legend.text = element_text(size = text_size),
legend.title = element_text(size = text_size),
plot.title = element_text(size=plot_title_size),
panel.border = element_rect(colour="black",fill=NA),
...)
}
#' @importFrom ggplot2 ggplot_gtable ggplot_build
extract_legend<-function(a.gplot){
tmp <- ggplot_gtable(ggplot_build(a.gplot))
leg <- which(sapply(tmp$grobs, function(x) x$name) == "guide-box")
legend <- tmp$grobs[[leg]]
return(legend)
}
#' @importFrom scales extended_breaks
histogram_breaks <-function(x){
default_breaks <- extended_breaks()(x)
if((max(default_breaks)-min(default_breaks)) < 0.01){
return(median(default_breaks))
}else{
return(default_breaks)
}
}
#' Plot bdrc model objects
#'
#' Visualize results from model objects in bdrc, plm0, plm, gplm0,gplm
#' @param x an object of class "plm0","plm","gplm0" or "gplm".
#' @param type a character denoting what type of plot should be drawn. Defaults to "rating_curve". Possible types are
#' \itemize{
#' \item{"rating_curve"}{ to plot the rating curve.}
#' \item{"rating_curve_mean"}{ to plot the posterior mean of the rating curve.}
#' \item{"f"}{ to plot the power-law exponent.}
#' \item{"beta"}{ to plot the random effect in the power-law exponent.}
#' \item{"sigma_eps"}{ to plot the standard deviation on the data level.}
#' \item{"residuals"}{ to plot the log residuals.}
#' \item{"trace"}{ to plot trace plots of parameters given in param.}
#' \item{"histogram"}{ to plot histograms of parameters given in param.}
#' \item{"panel"}{ to plot a 2x2 panel of plots: "rating curve", "residuals", "f" and "sigma_eps"}
#' }
#' @param param a character vector with the parameters to plot. Defaults to NULL and is only used if type is "trace" or "histogram". Allowed values are the parameters given in the model summary of x as well as "hyperparameters" or "latent_parameters" for specific groups of parameters.
#' @param transformed a logical value indicating whether the quantity should be plotted on a transformed scale used during the Bayesian inference. Defaults to FALSE.
#' @param title a character denoting the title of the plot. Defaults to NULL, i.e. no title.
#' @param xlim numeric vector of length 2, denoting the limits on the x axis of the plot. Applicable for types "rating_curve","rating_curve_mean","f","beta","sigma_eps","residuals".
#' @param ylim numeric vector of length 2, denoting the limits on the y axis of the plot. Applicable for types "rating_curve","rating_curve_mean","f","beta","sigma_eps","residuals".
#' @return returns an object of class ggplot2.
#' @importFrom ggplot2 ggplot aes geom_point geom_path geom_histogram geom_abline geom_hline geom_smooth facet_wrap scale_color_manual scale_x_continuous scale_y_continuous expansion label_parsed ggtitle xlab ylab geom_blank margin element_text theme
#' @importFrom rlang .data
#' @importFrom stats median
#' @keywords internal
plot_fun <- function(x,type='rating_curve',param=NULL,transformed=FALSE,title=NULL,xlim=NULL,ylim=NULL,...){
color_palette <- c("green","red","slateblue1","hotpink","#56B4E9","#E69F00","#000000","#999999","#CC79A7","#D55E00","#0072B2","#009E73")
legal_types <- c('rating_curve','rating_curve_mean','f','beta','sigma_eps','residuals','trace','histogram','r_hat','autocorrelation')
if(!(type %in% legal_types)){
stop(paste('Type argument not recognized. Possible types are:\n -',paste(legal_types,collapse='\n - ')))
}
mod_params <- get_param_names(class(x),c_param=x$run_info$c_param)
if(is.null(param)){
param <- mod_params #defaults to all stage independent model parameters
}else{
param <- get_args_rollout(param,mod_params)
}
if(type=='trace'){
plot_dat <- gather_draws(x,param,transformed=transformed)
if('h' %in% names(plot_dat)){
stop('Plots of type "trace" can only be of stage-independent parameters')
}
params <- unique(plot_dat$name)
if(length(params)>1){
param_levels <- get_parameter_levels(params)
plot_dat$name_expr <- factor(plot_dat$name,levels=param_levels,labels=sapply(param_levels,get_param_expression))
plot_dat$chain <- factor(as.character(plot_dat$chain),levels=1:max(plot_dat$chain))
p <- ggplot(plot_dat,aes(x=.data$iter,y=.data$value,col=.data$chain)) +
geom_path(alpha=0.7,linewidth=0.3) +
facet_wrap(~name_expr,scales='free',labeller = label_parsed) +
scale_color_manual(values=c("#BC3C29FF","#0072B5FF","#E18727FF","#20854EFF"),
name='Chain number') +
xlab('Iteration') +
ylab('') +
ggtitle(if(!is.null(title)) title else "Traceplots") +
theme_bdrc()
}else{
param_expr <- get_param_expression(params)
plot_dat$chain_name <- paste0('Chain nr ',plot_dat$chain)
p <- ggplot(plot_dat,aes(x=.data$iter,y=.data$value)) +
geom_path(col="#0072B5FF",linewidth=0.3) +
facet_wrap(~chain_name,scales='free') +
xlab('Iteration') +
ylab(parse(text=param_expr)) +
ggtitle(if(!is.null(title)) title else "Traceplots") +
theme_bdrc()
}
}else if(type=='histogram'){
plot_dat <- gather_draws(x,param,transformed=transformed)
if('h' %in% names(plot_dat)){
stop('Plots of type "histogram" can only be of stage-independent parameters')
}
params <- unique(plot_dat$name)
param_levels <- get_parameter_levels(params)
plot_dat$name_expr <- factor(plot_dat$name,levels=param_levels,labels=sapply(param_levels,get_param_expression))
plot_dat$chain <- factor(as.character(plot_dat$chain),levels=1:max(plot_dat$chain))
p <- ggplot(plot_dat,aes(x=.data$value)) +
geom_histogram(bins=50,fill="#0072B5FF") +
facet_wrap(~name_expr,scales='free',labeller=label_parsed,strip.position = "bottom") +
scale_x_continuous(breaks=histogram_breaks) +
scale_y_continuous(expand=c(0,0,0.05,0)) +
xlab('') +
ylab('') +
ggtitle(if(!is.null(title)) title else if( length(param)==1 ) "Histogram of posterior draws" else "Histograms of posterior draws" ) +
theme_bdrc() +
theme(plot.title = element_text( vjust = 2 ),
strip.placement = "outside",
plot.margin = ggplot2::margin(t = 10, r = 10, b = -10, l = 0, unit = "pt"))
}else if(type=='rating_curve' | type=='rating_curve_mean'){
if(transformed){
x_lab <- "paste('','',log,,,,'(','',italic(paste('h-',hat(paste('c')))),')','','')"
y_lab <- "paste('','',log,,,,'(','',italic(paste('Q')),')','','')"
c_hat <- if(is.null(x$run_info$c_param)) median(x$c_posterior) else x$run_info$c_param
h_min <- min(x$data[[all.vars(x$formula)[2]]])
plot_dat <- merge(x[[type]][x[[type]]$h>=h_min,],x$data,by.x='h',by.y=all.vars(x$formula)[2],all.x=TRUE)
plot_dat[,'log(h-c_hat)'] <- log(plot_dat$h-c_hat)
plot_dat$log_Q <- log(plot_dat[, all.vars(x$formula)[1]])
plot_dat$log_lower <- log(plot_dat$lower)
plot_dat$log_median <- log(plot_dat$median)
plot_dat$log_upper <- log(plot_dat$upper)
p <- ggplot(data=plot_dat) +
geom_path(aes(x=.data$`log(h-c_hat)`,y=.data$log_median),alpha=0.95) +
geom_path(aes(x=.data$`log(h-c_hat)`,y=.data$log_lower),linetype='dashed',alpha=0.95) +
geom_path(aes(x=.data$`log(h-c_hat)`,y=.data$log_upper),linetype='dashed',alpha=0.95) +
geom_point(data=plot_dat[!is.na(plot_dat$log_Q),],aes(x=.data$`log(h-c_hat)`,y=.data$log_Q), size=.9, shape=21, fill="gray60", color="black",alpha=0.95) +
scale_x_continuous(limits=if(!is.null(xlim)) xlim else c(NA,NA),expand=c(0.01,0)) +
scale_y_continuous(limits=if(!is.null(ylim)) ylim else c(NA,NA),expand=c(0.01,0)) +
xlab(parse(text=x_lab)) +
ylab(parse(text=y_lab)) +
ggtitle(if(!is.null(title)) title else "Log-transformed rating curve") +
theme_bdrc() +
theme(plot.title = element_text( vjust = 2 ))
}else{
x_lab <- "paste('','',italic(paste('Q')),paste('['),italic(paste('m',phantom() ^ {paste('3')},'/s')),paste(']'),'')"
y_lab <- "paste('','',italic(paste('h')),paste('['),italic(paste('m')),paste(']'),'')"
p <- ggplot(data=x[[type]]) +
geom_path(aes(x=.data$median,y=.data$h),alpha=0.95) +
geom_path(aes(x=.data$lower,y=.data$h),linetype='dashed',alpha=0.95) +
geom_path(aes(x=.data$upper,y=.data$h),linetype='dashed',alpha=0.95) +
geom_point(data=x$data,aes(.data[[all.vars(x$formula)[1]]],.data[[all.vars(x$formula)[2]]]), size=.9, shape=21, fill="gray60", color="black",alpha=0.95) +
scale_x_continuous(limits=if(!is.null(xlim)) xlim else c(NA,NA),expand=expansion(mult=0.01)) +
#scale_x_continuous(limits=if(!is.null(xlim)) xlim else c(0,max(x$rating_curve$upper,x$data$Q)),expand=c(0.01,0)) +
scale_y_continuous(limits=if(!is.null(ylim)) ylim else c(NA,NA),expand=c(0.01,0)) +
xlab(parse(text=x_lab)) +
ylab(parse(text=y_lab)) +
ggtitle(if(!is.null(title)) title else "Rating curve") +
theme_bdrc() +
theme(plot.title = element_text( vjust = 2 ))
}
}else if(type=='sigma_eps'){
x_lab <- "paste('','',italic(paste('h')),paste('['),italic(paste('m')),paste(']'),'')"
h_in_data <- x$data[,all.vars(x$formula)[2],drop=TRUE]
if('sigma_eps_summary' %in% names(x)){
y_lab <- "paste('','',sigma,,,,phantom() [ {paste('',epsilon,,,)} ],'(','',italic(paste('h')),')','','')"
plot_dat <- x$sigma_eps_summary[x$sigma_eps_summary$h>=min(h_in_data) & x$sigma_eps_summary$h<=max(h_in_data),]
}else{
y_lab <- "paste('','',sigma,,,,phantom() [ {paste('',epsilon,,,)} ],'')"
plot_dat <- data.frame(h=x$data[,all.vars(x$formula)[2],drop=TRUE],
lower=x$param_summary['sigma_eps','lower'],
median=x$param_summary['sigma_eps','median'],
upper=x$param_summary['sigma_eps','upper'])
}
p <- ggplot(data=plot_dat) +
geom_path(aes(x=.data$h,y=.data$median)) +
geom_path(aes(x=.data$h,y=.data$lower),linetype='dashed') +
geom_path(aes(x=.data$h,y=.data$upper),linetype='dashed') +
xlab(parse(text=x_lab)) +
ylab(parse(text=y_lab)) +
scale_x_continuous(limits=if(!is.null(xlim)) xlim else c(NA,NA),expand=c(0,0)) +
scale_y_continuous(limits=if(!is.null(ylim)) ylim else c(0,max(plot_dat$upper)*1.1),expand=c(0,0)) +
ggtitle(if(!is.null(title)) title else "Std. dev. of the error terms") +
theme_bdrc() +
theme(plot.title = element_text( vjust = 2 ))
}else if(type=='beta'){
if(!('beta_summary' %in% names(x))){
stop('Plots of type "beta" are only for models with stage dependent power law exponent, s.a. "gplm0" and "gplm"')
}
x_lab <- "paste('','',italic(paste('h')),paste('['),italic(paste('m')),paste(']'),'')"
y_lab <- "paste('','',beta,,,,'(','',italic(paste('h')),')','','')"
h_in_data <- x$data[,all.vars(x$formula)[2],drop=TRUE]
p <- ggplot(data=x$beta_summary[x$beta_summary$h>=min(h_in_data) & x$beta_summary$h<=max(h_in_data),]) +
geom_path(aes(.data$h,.data$median)) +
geom_path(aes(.data$h,.data$lower),linetype='dashed') +
geom_path(aes(.data$h,.data$upper),linetype='dashed') +
xlab(parse(text=x_lab)) +
ylab(parse(text=y_lab)) +
scale_x_continuous(if(!is.null(xlim)) xlim else c(NA,NA),expand=c(0,0)) +
scale_y_continuous(limits=if(!is.null(ylim)) ylim else c(NA,NA),expand=expansion(mult=rep(.05,2))) +
ggtitle(if(!is.null(title)) title else "Power-law exponent deviations") +
theme_bdrc() +
theme(plot.title = element_text( vjust = 2 ))
}else if(type=='f'){
x_lab <- "paste('','',italic(paste('h')),paste('['),italic(paste('m')),paste(']'),'')"
h_in_data <- x$data[,all.vars(x$formula)[2],drop=TRUE]
if('f_summary' %in% names(x)){
y_lab <- "paste('','',italic(paste('b+',beta,,,,'(','h',')','')),'')"
plot_dat <- x$f_summary[x$f_summary$h>=min(h_in_data) & x$f_summary$h<=max(h_in_data),]
}else{
y_lab <- "paste('','',italic(paste('b')),'')"
plot_dat <- data.frame(h=x$data[,all.vars(x$formula)[2],drop=TRUE],
lower=x$param_summary['b','lower'],
median=x$param_summary['b','median'],
upper=x$param_summary['b','upper'])
}
p <- ggplot(data=plot_dat) +
geom_path(aes(.data$h,.data$median)) +
geom_path(aes(.data$h,.data$lower),linetype='dashed') +
geom_path(aes(.data$h,.data$upper),linetype='dashed') +
xlab(parse(text=x_lab)) +
ylab(parse(text=y_lab)) +
scale_x_continuous(limits=if(!is.null(xlim)) xlim else c(NA,NA),expand=c(0,0)) +
scale_y_continuous(limits=if(!is.null(ylim)) ylim else c(min(1,0.9*min(plot_dat$lower)),max(3.5,1.1*max(plot_dat$upper))),expand=c(0,0)) +
ggtitle(if(!is.null(title)) title else "Power-law exponent") +
theme_bdrc() +
theme(plot.title = element_text( vjust = 2 ))
}else if(type=='residuals'){
resid_dat <- get_residuals_dat(x)
y_lab <- "paste('','log','(','',italic(paste('Q')),')','','-log','(','',italic(paste('',hat(paste('Q')))),')','','')"
x_lab <- "paste('','log','(','',italic(paste('h',phantom() - phantom(),'',hat(paste('c')))),')','','')"
method <- 'loess'
span <- 0.3
p <- ggplot(data=resid_dat) +
geom_hline(yintercept=0,linewidth=0.8,alpha=.95) +
geom_point(data=resid_dat[!is.na(resid_dat$Q),],aes(.data$`log(h-c_hat)`,.data$r_median), size=.9, shape=21, fill="gray60", color="black",alpha=0.95) +
geom_blank(aes(y=-.data$r_median)) +
geom_blank(aes(y=-.data$r_upper)) +
geom_blank(aes(y=-.data$r_lower)) +
geom_blank(aes(y=-.data$m_upper)) +
geom_blank(aes(y=-.data$m_lower)) +
geom_smooth(aes(x=.data$`log(h-c_hat)`,y=.data$r_upper),span=span,se=FALSE,stat = "smooth",color='black',linetype='dashed',linewidth=0.5,alpha=0.95,method=method,formula='y~x') +
geom_smooth(aes(x=.data$`log(h-c_hat)`,y=.data$r_lower),span=span,se=FALSE,stat = "smooth",color='black',linetype='dashed',linewidth=0.5,alpha=0.95,method=method,formula='y~x') +
geom_smooth(aes(x=.data$`log(h-c_hat)`,y=.data$m_upper),span=span,se=FALSE,stat = "smooth",color='black',linetype='solid',linewidth=0.3,alpha=0.95,method=method,formula='y~x') +
geom_smooth(aes(x=.data$`log(h-c_hat)`,y=.data$m_lower),span=span,se=FALSE,stat = "smooth",color='black',linetype='solid',linewidth=0.3,alpha=0.95,method=method,formula='y~x') +
xlab(parse(text=x_lab)) +
ylab(parse(text=y_lab)) +
scale_x_continuous(limits=if(!is.null(xlim)) xlim else c(NA,NA),expand=expansion(mult=rep(.01,2))) +
scale_y_continuous(limits=if(!is.null(ylim)) ylim else c(NA,NA),expand=expansion(mult=rep(.05,2))) +
ggtitle(if(!is.null(title)) title else "Residuals") +
theme_bdrc() +
theme(plot.title = element_text( vjust = 2 ))
}else if(type=='r_hat'){
rhat_dat <- get_rhat_dat(x,param)
rhat_dat$Rhat[rhat_dat$Rhat<1] <- 1
rhat_dat$Rhat[rhat_dat$Rhat>2] <- 2
param_expr <- parse(text=get_param_expression(param))
y_lab <- "paste('','',italic(paste('',hat(paste('R')))),'')"
p <- ggplot(data=rhat_dat, aes(x=.data$iterations,y=.data$Rhat,color=.data$parameters)) +
geom_hline(yintercept=1.1,linetype='dashed') +
geom_line(na.rm=TRUE) +
scale_x_continuous(limits=c(4*x$run_info$thin+x$run_info$burnin,x$run_info$nr_iter),breaks=c(5000,10000,15000),expand=c(0,0)) +
scale_y_continuous(limits=c(1,2),breaks=c(1,1.1,1.2,1.4,1.6,1.8,2),expand=c(0,0)) +
scale_color_manual(values=color_palette,name=class(x),labels=param_expr) +
xlab('Iteration') +
ylab(parse(text=y_lab)) +
ggtitle(if(!is.null(title)) title else "Gelman-Rubin statistic") +
theme_bdrc() +
theme(plot.title = element_text( vjust = 2 ),
axis.title.y = element_text( vjust = 3 ),
plot.margin = ggplot2::margin( t=7, r=7, b=7, l=12, unit = "pt" ) )
}else if(type=='autocorrelation'){
auto_dat <- do.call('rbind',lapply(param,function(p) data.frame(lag=x$autocorrelation$lag,param=p,corr=x$autocorrelation[,p])))
param_expr <- parse(text=get_param_expression(param))
max_lag <- nrow(x$autocorrelation)
p <- ggplot(data=auto_dat, aes(x=.data$lag,y=.data$corr,color=.data$param)) +
geom_hline(yintercept=0) +
geom_line() +
geom_point(size=1) +
scale_x_continuous(limits=c(1,max_lag),labels=c(1,seq(5,max_lag,5)),breaks=c(1,seq(5,max_lag,5)),expand=c(0,1)) +
scale_y_continuous(limits=c(min(auto_dat$corr,-1/11),1),expand=c(0,0)) +
scale_color_manual(values=color_palette,name=class(x),labels=param_expr) +
xlab('Lag') +
ylab('Sample autocorrelation') +
ggtitle(if(!is.null(title)) title else "Autocorrelation in posterior draws") +
theme_bdrc() +
theme(plot.title = element_text( vjust = 2 ),
axis.title.y = element_text( vjust = 3 ),
plot.margin = ggplot2::margin( t=7, r=7, b=7, l=12, unit = "pt" ) )
}
return(p)
}
#' @importFrom gridExtra arrangeGrob
#' @importFrom grid textGrob gpar unit unit.pmax
#' @importFrom ggplot2 theme guides guide_legend ggplotGrob
plot_grob <- function(x,type,transformed=FALSE){
if(type=='panel'){
panel_types <- c('rating_curve','residuals','f','sigma_eps')
grob_list <- lapply(panel_types,function(ty){
ggplotGrob(plot_fun(x,type=ty,transformed=transformed))
})
maxHeight <- unit.pmax( grob_list[[1]]$heights[2:9], grob_list[[2]]$heights[2:9],
grob_list[[3]]$heights[2:9], grob_list[[4]]$heights[2:9])
maxWidth <- unit.pmax( grob_list[[1]]$widths[2:5], grob_list[[2]]$widths[2:5],
grob_list[[3]]$widths[2:5], grob_list[[4]]$widths[2:5])
for(j in 1:4){
grob_list[[j]]$heights[2:9] <- as.list(maxHeight)
grob_list[[j]]$widths[2:5] <- as.list(maxWidth)
}
p <- do.call(arrangeGrob,c(grob_list,ncol=round(sqrt(length(panel_types)))))
}else if(type=='convergence_diagnostics'){
autocorrelation_plot <- plot_fun(x,type='autocorrelation') +
theme_bdrc(legend.key.size = unit(0.8, "lines"),
legend.justification = "top",
legend.key.width = unit(0.5,"cm"),
scaling=0.82)
r_hat_plot <- plot_fun(x,type='r_hat') +
theme_bdrc(legend.key.size = unit(0.8, "lines"),
legend.justification = "top",
legend.key.width = unit(1,"cm"),
scaling=0.82)
legend <- extract_legend(autocorrelation_plot)
p <- arrangeGrob(arrangeGrob(r_hat_plot+theme(legend.position="none"),
autocorrelation_plot+theme(legend.position="none"),nrow=1),
legend,ncol=2,widths=c(4,1))
}
return(p)
}
#' @importFrom stats approx
predict_fun <- function(object,newdata=NULL,wide=FALSE){
if(wide){
if(!is.null(newdata)){
stop('newdata must be NULL when wide is TRUE.')
}else{
newdata <- seq(ceiling(min(object$rating_curve$h)*100)/100,floor(max(object$rating_curve$h)*100)/100,by=0.01)
}
}else{
if(is.null(newdata)){
newdata <- object$rating_curve$h
}
}
if(!inherits(newdata,'numeric')){
stop('newdata must be a vector of type "numeric" or NULL')
}
if(any(is.na(newdata))){
stop('newdata must not include NA')
}
if(any(newdata>max(object$rating_curve$h))){
stop('newdata must contain values within the range of stage values used to fit the rating curve. See "h_max" option to extrapolate the rating curve to higher stages')
}
lower_pred <- approx(object$rating_curve$h,object$rating_curve$lower,xout=newdata)$y
median_pred <- approx(object$rating_curve$h,object$rating_curve$median,xout=newdata)$y
upper_pred <- approx(object$rating_curve$h,object$rating_curve$upper,xout=newdata)$y
pred_dat <- data.frame(h=newdata,lower=lower_pred,median=median_pred,upper=upper_pred)
pred_dat[is.na(pred_dat)] <- 0
if(wide){
pred_dat <- predict_wider(pred_dat)
}
return(pred_dat)
}
#' Print method for discharge rating curves
#'
#' Print a discharge rating curve model object
#' @param x an object of class "plm0", "plm", "gplm0" or "gplm".
#' @param ... not used in this function
#' @seealso \code{\link{plm0}}, \code{\link{plm}}, \code{\link{gplm0}}, \code{\link{gplm}} for fitting a discharge rating curve and \code{\link{summary.plm0}}, \code{\link{summary.plm}}, \code{\link{summary.gplm0}} and \code{\link{summary.gplm}} for summaries. It is also useful to look at \code{\link{plot.plm0}}, \code{\link{plot.plm}}, \code{\link{plot.gplm0}} and \code{\link{plot.gplm}} to help visualize all aspects of the fitted discharge rating curve. Additionally, \code{\link{spread_draws}} and \code{\link{spread_draws}} help working directly with the MCMC samples.
#' @describeIn print.plm0 Print method for plm0
#' @export
print.plm0 <- function(x,...){
print_fun(x)
}
#' Summary method for discharge rating curves
#'
#' Summarize a discharge rating curve model object
#' @param object an object of class "plm0", "plm", "gplm0" or "gplm".
#' @param ... Not used for this function
#' @seealso \code{\link{plm0}}, \code{\link{plm}}, \code{\link{gplm0}} and \code{\link{gplm}} for fitting a discharge rating curve. It is also useful to look at \code{\link{plot.plm0}}, \code{\link{plot.plm}}, \code{\link{plot.gplm0}} and \code{\link{plot.gplm}} to help visualize all aspects of the fitted discharge rating curve. Additionally, \code{\link{spread_draws}} and \code{\link{spread_draws}} help working directly with the MCMC samples.
#' @examples
#' \donttest{
#' data(krokfors)
#' set.seed(1)
#' plm0.fit <- plm0(formula=Q~W,data=krokfors,num_cores=2)
#' summary(plm0.fit)
#' }
#' @describeIn summary.plm0 Summary method for plm0
#' @export
summary.plm0 <- function(object,...){
summary_fun(object)
}
#' Autoplot method for discharge rating curves
#'
#' Visualize discharge rating curve model objects
#' @param object an object of class "plm0","plm","gplm0" or "gplm".
#' @param ... other plotting parameters (not used in this function)
#' @param type a character denoting what type of plot should be drawn. Defaults to "rating_curve". Possible types are
#' \itemize{
#' \item{"rating_curve"}{ to plot the rating curve.}
#' \item{"rating_curve_mean"}{ to plot the posterior mean of the rating curve.}
#' \item{"f"}{ to plot the power-law exponent.}
#' \item{"beta"}{ to plot the random effect in the power-law exponent.}
#' \item{"sigma_eps"}{ to plot the standard deviation on the data level.}
#' \item{"residuals"}{ to plot the log residuals.}
#' \item{"trace"}{ to plot trace plots of parameters given in param.}
#' \item{"histogram"}{ to plot histograms of parameters given in param.}
#' }
#' @param param a character vector with the parameters to plot. Defaults to NULL and is only used if type is "trace" or "histogram". Allowed values are the parameters given in the model summary of x as well as "hyperparameters" or "latent_parameters" for specific groups of parameters.
#' @param transformed a logical value indicating whether the quantity should be plotted on a transformed scale used during the Bayesian inference. Defaults to FALSE.
#' @param title a character denoting the title of the plot
#' @param xlim numeric vector of length 2, denoting the limits on the x axis of the plot. Applicable for types "rating_curve","rating_curve_mean","f","beta","sigma_eps","residuals".
#' @param ylim numeric vector of length 2, denoting the limits on the y axis of the plot. Applicable for types "rating_curve","rating_curve_mean","f","beta","sigma_eps","residuals".
#'
#' @return returns an object of class "ggplot2".
#' @seealso \code{\link{plm0}}, \code{\link{plm}}, \code{\link{gplm0}} and \code{\link{gplm}} for fitting a discharge rating curve and \code{\link{summary.plm0}}, \code{\link{summary.plm}}, \code{\link{summary.gplm0}} and \code{\link{summary.gplm}} for summaries. It is also useful to look at \code{\link{spread_draws}} and \code{\link{gather_draws}} to work directly with the MCMC samples.
#' @examples
#' \donttest{
#' library(ggplot2)
#' data(krokfors)
#' set.seed(1)
#' plm0.fit <- plm0(Q~W,krokfors,num_cores=2)
#' autoplot(plm0.fit)
#' autoplot(plm0.fit,transformed=TRUE)
#' autoplot(plm0.fit,type='histogram',param='c')
#' autoplot(plm0.fit,type='histogram',param='c',transformed=TRUE)
#' autoplot(plm0.fit,type='histogram',param='hyperparameters')
#' autoplot(plm0.fit,type='histogram',param='latent_parameters')
#' autoplot(plm0.fit,type='residuals')
#' autoplot(plm0.fit,type='f')
#' autoplot(plm0.fit,type='sigma_eps')
#' }
#' @describeIn autoplot.plm0 Autoplot method for plm0
#' @importFrom ggplot2 autoplot
#' @export
autoplot.plm0 <- function(object,...,type='rating_curve',param=NULL,transformed=FALSE,title=NULL,xlim=NULL,ylim=NULL){
plot_fun(object,type=type,param=param,transformed=transformed,title=title,xlim=xlim,ylim=ylim)
}
#' Plot method for discharge rating curves
#'
#' Visualize discharge rating curve model objects
#' @param x object of class "plm0", "plm", "gplm0" or "gplm".
#' @param ... other plotting parameters (not used in this function)
#' @param type a character denoting what type of plot should be drawn. Defaults to "rating_curve". Possible types are
#' \itemize{
#' \item{"rating_curve"}{ to plot the rating curve.}
#' \item{"rating_curve_mean"}{ to plot the posterior mean of the rating curve.}
#' \item{"f"}{ to plot the power-law exponent.}
#' \item{"beta"}{ to plot the random effect in the power-law exponent.}
#' \item{"sigma_eps"}{ to plot the standard deviation on the data level.}
#' \item{"residuals"}{ to plot the log residuals.}
#' \item{"trace"}{ to plot trace plots of parameters given in param.}
#' \item{"histogram"}{ to plot histograms of parameters given in param.}
#' \item{"panel"}{ to plot a 2x2 panel of plots: "rating curve", "residuals", "f" and "sigma_eps"}
#' }
#' @param param a character vector with the parameters to plot. Defaults to NULL and is only used if type is "trace" or "histogram". Allowed values are the parameters given in the model summary of x as well as "hyperparameters" or "latent_parameters" for specific groups of parameters.
#' @param transformed a logical value indicating whether the quantity should be plotted on a transformed scale used during the Bayesian inference. Defaults to FALSE.
#' @param title a character denoting the title of the plot
#' @param xlim numeric vector of length 2, denoting the limits on the x axis of the plot. Applicable for types "rating_curve","rating_curve_mean","f","beta","sigma_eps","residuals".
#' @param ylim numeric vector of length 2, denoting the limits on the y axis of the plot. Applicable for types "rating_curve","rating_curve_mean","f","beta","sigma_eps","residuals".
#'
#' @return No return value, called for side effects.
#' @seealso \code{\link{plm0}}, \code{\link{plm}}, \code{\link{gplm0}} and \code{\link{gplm}} for fitting a discharge rating curve and \code{\link{summary.plm0}}, \code{\link{summary.plm}}, \code{\link{summary.gplm0}} and \code{\link{summary.gplm}} for summaries. It is also useful to look at \code{\link{spread_draws}} and \code{\link{gather_draws}} to work directly with the MCMC samples.
#' @examples
#' \donttest{
#' data(krokfors)
#' set.seed(1)
#' plm0.fit <- plm0(formula=Q~W,data=krokfors,num_cores=2)
#'
#' plot(plm0.fit)
#' plot(plm0.fit,transformed=TRUE)
#' plot(plm0.fit,type='histogram',param='c')
#' plot(plm0.fit,type='histogram',param='c',transformed=TRUE)
#' plot(plm0.fit,type='histogram',param='hyperparameters')
#' plot(plm0.fit,type='histogram',param='latent_parameters')
#' plot(plm0.fit,type='residuals')
#' plot(plm0.fit,type='f')
#' plot(plm0.fit,type='sigma_eps')
#' }
#' @describeIn plot.plm0 Plot method for plm0
#' @export
#' @importFrom grid grid.draw
#' @importFrom ggplot2 autoplot
plot.plm0 <- function(x,...,type='rating_curve',param=NULL,transformed=FALSE,title=NULL,xlim=NULL,ylim=NULL){
grob_types <- c('panel','convergence_diagnostics')
if(is.null(type) || !(type%in%grob_types)){
p <- autoplot(x,type=type,param=param,transformed=transformed,title=title,xlim=xlim,ylim=ylim)
print(p)
}else{
p <- plot_grob(x,type=type,transformed=transformed)
grid.draw(p)
}
}
#' Predict method for discharge rating curves
#'
#' Predict the discharge for given stage values based on a discharge rating curve model object.
#' @param object an object of class "plm0", "plm", "gplm0" or "gplm".
#' @param newdata a numeric vector of stage values for which to predict. If omitted, the stage values in the data are used.
#' @param wide a logical value denoting whether to produce a wide prediction output.If TRUE, then the output is a table with median prediction values for an equally spaced grid of stages with 1 cm increments, each row containing predictions in a decimeter range of stages.
#' @param ... not used in this function
#' @return an object of class "data.frame" with four columns, h (stage), lower (2.5\% posterior predictive quantile), median (50\% posterior predictive quantile), upper (97.5\% posterior predictive quantile). If wide=TRUE, a matrix as described above (see wide parameter) is returned.
#' @seealso \code{\link{plm0}}, \code{\link{plm}}, \code{\link{gplm0}} and \code{\link{gplm}} for fitting a discharge rating curve and \code{\link{summary.plm0}}, \code{\link{summary.plm}}, \code{\link{summary.gplm0}} and \code{\link{summary.gplm}} for summaries. It is also useful to look at \code{\link{plot.plm0}}, \code{\link{plot.plm}}, \code{\link{plot.gplm0}} and \code{\link{plot.gplm}} to help visualize all aspects of the fitted discharge rating curve. Additionally, \code{\link{spread_draws}} and \code{\link{spread_draws}} help working directly with the MCMC samples.
#' @examples
#' \donttest{
#' data(krokfors)
#' set.seed(1)
#' plm0.fit <- plm0(formula=Q~W,data=krokfors,h_max=10,num_cores=2)
#' #predict rating curve on a equally 10 cm spaced grid from 9 to 10 meters
#' predict(plm0.fit,newdata=seq(9,10,by=0.1))
#' }
#' @describeIn predict.plm0 Predict method for plm0
#' @export
predict.plm0 <- function(object,...,newdata=NULL,wide=FALSE){
predict_fun(object,newdata=newdata,wide=wide)
}
#' @describeIn print.plm0 Print method for plm
#' @export
#'
print.plm <- function(x,...){
print_fun(x)
}
#' @describeIn summary.plm0 Summary method for plm
#' @export
summary.plm <- function(object,...){
summary_fun(object)
}
#' @describeIn autoplot.plm0 Autoplot method for plm
#' @export
autoplot.plm <- function(object,...,type='rating_curve',param=NULL,transformed=FALSE,title=NULL,xlim=NULL,ylim=NULL){
plot_fun(object,type=type,param=param,transformed=transformed,title=title,xlim=xlim,ylim=ylim)
}
#' @describeIn plot.plm0 Plot method for plm
#' @export
#' @importFrom grid grid.draw
#' @importFrom ggplot2 autoplot
plot.plm <- function(x,...,type='rating_curve',param=NULL,transformed=FALSE,title=NULL,xlim=NULL,ylim=NULL){
grob_types <- c('panel','convergence_diagnostics')
if(is.null(type) || !(type%in%grob_types)){
p <- autoplot(x,type=type,param=param,transformed=transformed,title=title,xlim=xlim,ylim=ylim)
print(p)
}else{
p <- plot_grob(x,type=type,transformed=transformed)
grid.draw(p)
}
}
#' @describeIn predict.plm0 Predict method for plm
#' @export
predict.plm <- function(object,...,newdata=NULL,wide=FALSE){
predict_fun(object,newdata=newdata,wide=wide)
}
#' @describeIn print.plm0 Print method for gplm0
#' @export
print.gplm0 <- function(x,...){
print_fun(x)
}
#' @describeIn summary.plm0 Summary method for gplm0
#' @export
summary.gplm0 <- function(object,...){
summary_fun(object)
}
#' @describeIn autoplot.plm0 Autoplot method for gplm0
#' @export
autoplot.gplm0 <- function(object,...,type='rating_curve',param=NULL,transformed=FALSE,title=NULL,xlim=NULL,ylim=NULL){
plot_fun(object,type=type,param=param,transformed=transformed,title=title,xlim=xlim,ylim=ylim)
}
#' @describeIn plot.plm0 Plot method for gplm0
#' @export
#' @importFrom grid grid.draw
#' @importFrom ggplot2 autoplot
plot.gplm0 <- function(x,...,type='rating_curve',param=NULL,transformed=FALSE,title=NULL,xlim=NULL,ylim=NULL){
grob_types <- c('panel','convergence_diagnostics')
if(is.null(type) || !(type%in%grob_types)){
p <- autoplot(x,type=type,param=param,transformed=transformed,title=title,xlim=xlim,ylim=ylim)
print(p)
}else{
p <- plot_grob(x,type=type,transformed=transformed)
grid.draw(p)
}
}
#' @describeIn predict.plm0 Predict method for gplm0
#' @export
predict.gplm0 <- function(object,...,newdata=NULL,wide=FALSE){
predict_fun(object,newdata=newdata,wide=wide)
}
#' @describeIn print.plm0 Print method for gplm
#' @export
print.gplm <- function(x,...){
print_fun(x)
}
#' @describeIn summary.plm0 Summary method for gplm
#' @export
summary.gplm <- function(object,...){
summary_fun(object)
}
#' @describeIn autoplot.plm0 Autoplot method for gplm
#' @export
autoplot.gplm <- function(object,...,type='rating_curve',param=NULL,transformed=FALSE,title=NULL,xlim=NULL,ylim=NULL){
plot_fun(object,type=type,param=param,transformed=transformed,title=title,xlim=xlim,ylim=ylim)
}
#' @describeIn plot.plm0 Plot method for gplm
#' @export
#' @importFrom grid grid.draw
#' @importFrom ggplot2 autoplot
plot.gplm <- function(x,...,type='rating_curve',param=NULL,transformed=FALSE,title=NULL,xlim=NULL,ylim=NULL){
grob_types <- c('panel','convergence_diagnostics')
if(is.null(type) || !(type%in%grob_types)){
p <- autoplot(x,type=type,param=param,transformed=transformed,title=title,xlim=xlim,ylim=ylim)
print(p)
}else{
p <- plot_grob(x,type=type,transformed=transformed)
grid.draw(p)
}
}
#' @describeIn predict.plm0 Predict method for gplm
#' @export
predict.gplm <- function(object,...,newdata=NULL,wide=FALSE){
predict_fun(object,newdata=newdata,wide=wide)
}