irf_nlVAR_TEST.Rout.save

R version 3.6.1 (2019-07-05) -- "Action of the Toes"
Copyright (C) 2019 The R Foundation for Statistical Computing
Platform: x86_64-pc-linux-gnu (64-bit)

R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.

  Natural language support but running in an English locale

R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.

Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.

> library(tsDyn)
> suppressMessages(library(dplyr))
> library(purrr)
> library(tidyr)
> select <- dplyr::select
> suppressWarnings(RNGversion("3.5.3"))
> 
> ############################
> ### Load data
> ############################
> 
> path_mod_multi <- system.file("testdata/models_multivariate.rds", package = "tsDyn")
> 
> models_multivariate <- readRDS(path_mod_multi)
> 
> models_multivariate %>% 
+   mutate(across(where(is.list), class)) %>% 
+   as.data.frame() %>% 
+   head(12)
   lag include model object object_vars nthresh
1    1    both   VAR   list        list      NA
2    1   const   VAR   list        list      NA
3    1    none   VAR   list        list      NA
4    1   trend   VAR   list        list      NA
5    2    both   VAR   list        list      NA
6    2   const   VAR   list        list      NA
7    2    none   VAR   list        list      NA
8    2   trend   VAR   list        list      NA
9    1    both  VECM   list        list      NA
10   1   const  VECM   list        list      NA
11   1    none  VECM   list        list      NA
12   1   trend  VECM   list        list      NA
> 
> 
> ############################
> ### VAR
> ############################
> 
> irf_any <- tsDyn:::irf_any
> irf_1 <- tsDyn:::irf_1
> irf_1.nlVar <- tsDyn:::irf_1.nlVar
> 
> ## manual comparisons
> mod_random_1 <- filter(models_multivariate, lag ==2)$object[[2]]
> mod_random_1_vars <- filter(models_multivariate, lag ==2)$object_vars[[2]]
> 
> irf_any(mod_random_1, boot = FALSE)$irf[[1]]
          dolcan        cpiUSA
 [1,] 0.01279642 -0.0048756343
 [2,] 0.01474258 -0.0003293138
 [3,] 0.01481730  0.0018693512
 [4,] 0.01458848  0.0018642930
 [5,] 0.01432107  0.0006853138
 [6,] 0.01405581 -0.0010343557
 [7,] 0.01379700 -0.0029806380
 [8,] 0.01354404 -0.0050076916
 [9,] 0.01329620 -0.0070495613
[10,] 0.01305302 -0.0090768908
[11,] 0.01281423 -0.0110768628
> irf(mod_random_1, boot = FALSE)$irf[[1]]
          dolcan        cpiUSA
 [1,] 0.01279642 -0.0048756343
 [2,] 0.01474258 -0.0003293138
 [3,] 0.01481730  0.0018693512
 [4,] 0.01458848  0.0018642930
 [5,] 0.01432107  0.0006853138
 [6,] 0.01405581 -0.0010343557
 [7,] 0.01379700 -0.0029806380
 [8,] 0.01354404 -0.0050076916
 [9,] 0.01329620 -0.0070495613
[10,] 0.01305302 -0.0090768908
[11,] 0.01281423 -0.0110768628
> irf(mod_random_1_vars, boot = FALSE)$irf[[1]]
          dolcan        cpiUSA
 [1,] 0.01279642 -0.0048756343
 [2,] 0.01474258 -0.0003293138
 [3,] 0.01481730  0.0018693512
 [4,] 0.01458848  0.0018642930
 [5,] 0.01432107  0.0006853138
 [6,] 0.01405581 -0.0010343557
 [7,] 0.01379700 -0.0029806380
 [8,] 0.01354404 -0.0050076916
 [9,] 0.01329620 -0.0070495613
[10,] 0.01305302 -0.0090768908
[11,] 0.01281423 -0.0110768628
> 
> irf_any(mod_random_1, boot = FALSE, ortho = FALSE)$irf[[1]]
         dolcan      cpiUSA
 [1,] 1.0000000  0.00000000
 [2,] 1.1506724  0.52942976
 [3,] 1.1556807  0.77998598
 [4,] 1.1374441  0.81507635
 [5,] 1.1164438  0.73894786
 [6,] 1.0957289  0.61180990
 [7,] 1.0755711  0.46302596
 [8,] 1.0558941  0.30615552
 [9,] 1.0366258  0.14732251
[10,] 1.0177245 -0.01074405
[11,] 0.9991672 -0.16684746
> irf(mod_random_1, boot = FALSE, ortho = FALSE)$irf[[1]]
         dolcan      cpiUSA
 [1,] 1.0000000  0.00000000
 [2,] 1.1506724  0.52942976
 [3,] 1.1556807  0.77998598
 [4,] 1.1374441  0.81507635
 [5,] 1.1164438  0.73894786
 [6,] 1.0957289  0.61180990
 [7,] 1.0755711  0.46302596
 [8,] 1.0558941  0.30615552
 [9,] 1.0366258  0.14732251
[10,] 1.0177245 -0.01074405
[11,] 0.9991672 -0.16684746
> irf(mod_random_1_vars, boot = FALSE, ortho = FALSE)$irf[[1]]
         dolcan      cpiUSA
 [1,] 1.0000000  0.00000000
 [2,] 1.1506724  0.52942976
 [3,] 1.1556807  0.77998598
 [4,] 1.1374441  0.81507635
 [5,] 1.1164438  0.73894786
 [6,] 1.0957289  0.61180990
 [7,] 1.0755711  0.46302596
 [8,] 1.0558941  0.30615552
 [9,] 1.0366258  0.14732251
[10,] 1.0177245 -0.01074405
[11,] 0.9991672 -0.16684746
> 
> ### irf _1
> models_IRF_1 <- models_multivariate %>% 
+   filter(model == "VAR") %>% 
+   mutate(irf = map(object, ~irf_1(.)))
> 
> models_IRF_1$irf %>% 
+   bind_rows() %>% 
+   head() %>% 
+   print(digits=3)
  dolcan cpiUSA impulse
1  1.000  0.000  dolcan
2  0.987 -0.140  dolcan
3  0.974 -0.278  dolcan
4  0.962 -0.414  dolcan
5  0.949 -0.549  dolcan
6  0.937 -0.683  dolcan
> 
> ### irf_any
> # irf.NULL <- function(x) NULL
> # irf.ca.jo <- function(x) irf(vec2var(ca.jo))
> 
> models_VAR <- models_multivariate %>% 
+   filter(model == "VAR")
> 
> ## older method
> models_IRF_any <- models_multivariate %>% 
+   filter(model == "VAR") %>% 
+   mutate(ortho = list(tibble(ortho =c(TRUE, FALSE)))) %>% 
+   unnest(., ortho) %>% 
+   mutate(irf = map2(object, ortho, ~irf_any(.x,  boot = TRUE, runs = 1, seed = 7, ortho = .y)),
+          irf_vars = map2(object_vars, ortho, ~irf(.x, runs = 1, seed = 7, ortho = .y)),
+          irf_vec2 = map2(object, ortho, ~irf(.x,  boot = FALSE, runs = 1, seed = 7, ortho = .y)))
> 
> models_IRF_any %>% 
+   mutate(across(where(is.list), class)) %>% 
+   as.data.frame()
   lag include model object object_vars nthresh ortho  irf irf_vars irf_vec2
1    1    both   VAR   list        list      NA  TRUE list     list     list
2    1    both   VAR   list        list      NA FALSE list     list     list
3    1   const   VAR   list        list      NA  TRUE list     list     list
4    1   const   VAR   list        list      NA FALSE list     list     list
5    1    none   VAR   list        list      NA  TRUE list     list     list
6    1    none   VAR   list        list      NA FALSE list     list     list
7    1   trend   VAR   list        list      NA  TRUE list     list     list
8    1   trend   VAR   list        list      NA FALSE list     list     list
9    2    both   VAR   list        list      NA  TRUE list     list     list
10   2    both   VAR   list        list      NA FALSE list     list     list
11   2   const   VAR   list        list      NA  TRUE list     list     list
12   2   const   VAR   list        list      NA FALSE list     list     list
13   2    none   VAR   list        list      NA  TRUE list     list     list
14   2    none   VAR   list        list      NA FALSE list     list     list
15   2   trend   VAR   list        list      NA  TRUE list     list     list
16   2   trend   VAR   list        list      NA FALSE list     list     list
> 
> ## showquick summary
> irf_extract_here <- function(x) {
+   head(x$irf[[1]], 2) %>% 
+     as.data.frame() %>% 
+     mutate(type = "irf") %>% 
+     rbind(head(x$Upper[[1]], 2) %>% 
+             as.data.frame() %>% 
+             mutate(type = "Upper_CI")) %>% 
+     relocate(type)
+ }
> 
> ## show head of irf any
> map_dfr(models_IRF_any$irf, irf_extract_here) %>% 
+   as.data.frame() %>% 
+   head(10)%>% 
+   mutate(across(where(is.numeric), ~round(., 6)))
       type   dolcan    cpiUSA
1       irf 0.012960 -0.008008
2       irf 0.012792 -0.009835
3  Upper_CI 0.000000  0.168830
4  Upper_CI 0.000284  0.168505
5       irf 1.000000  0.000000
6       irf 0.987108 -0.139549
7  Upper_CI 0.000000  1.000000
8  Upper_CI 0.001679  0.998075
9       irf 0.012941 -0.007913
10      irf 0.012773 -0.009736
> 
> 
> ## compare with vars
> all.equal(models_IRF_any$irf[[1]]$irf, 
+           models_IRF_any$irf_vars[[1]]$irf)
[1] TRUE
> models_IRF_any$irf[[1]]$irf[[1]]
          dolcan       cpiUSA
 [1,] 0.01296044 -0.008008276
 [2,] 0.01279219 -0.009835144
 [3,] 0.01262584 -0.011642696
 [4,] 0.01246137 -0.013431155
 [5,] 0.01229877 -0.015200739
 [6,] 0.01213800 -0.016951666
 [7,] 0.01197905 -0.018684149
 [8,] 0.01182189 -0.020398399
 [9,] 0.01166652 -0.022094627
[10,] 0.01151290 -0.023773038
[11,] 0.01136101 -0.025433838
> models_IRF_any$irf_vars[[1]]$irf[[1]]
          dolcan       cpiUSA
 [1,] 0.01296044 -0.008008276
 [2,] 0.01279219 -0.009835144
 [3,] 0.01262584 -0.011642696
 [4,] 0.01246137 -0.013431155
 [5,] 0.01229877 -0.015200739
 [6,] 0.01213800 -0.016951666
 [7,] 0.01197905 -0.018684149
 [8,] 0.01182189 -0.020398399
 [9,] 0.01166652 -0.022094627
[10,] 0.01151290 -0.023773038
[11,] 0.01136101 -0.025433838
> models_IRF_any$irf_vec2[[1]]$irf[[1]]
          dolcan       cpiUSA
 [1,] 0.01296044 -0.008008276
 [2,] 0.01279219 -0.009835144
 [3,] 0.01262584 -0.011642696
 [4,] 0.01246137 -0.013431155
 [5,] 0.01229877 -0.015200739
 [6,] 0.01213800 -0.016951666
 [7,] 0.01197905 -0.018684149
 [8,] 0.01182189 -0.020398399
 [9,] 0.01166652 -0.022094627
[10,] 0.01151290 -0.023773038
[11,] 0.01136101 -0.025433838
> 
> comp <- models_IRF_any %>% 
+   mutate(comp_irf_tsD_vars = map2(irf, irf_vars,  ~all.equal(.x$irf, .y$irf)),
+          is_same = map_lgl(comp_irf_tsD_vars, ~isTRUE(.)),
+          comp_irf_tsDOld_vars = map2(irf_vec2, irf_vars,  ~all.equal(.x$irf, .y$irf)),
+          is_same_tssDvec2 = map_lgl(comp_irf_tsDOld_vars, ~isTRUE(.)),
+          comp_irf_tsDOld_tsDNew = map2_lgl(irf, irf_vec2,  ~all.equal(.x$irf, .y$irf)),
+          is_same_tsD_2ver = map_lgl(comp_irf_tsDOld_tsDNew, ~isTRUE(.))) %>% 
+   dplyr::select(-starts_with("irf"), -starts_with("comp_irf"), comp_irf_tsDOld_tsDNew)
> 
> comp %>% 
+   dplyr::select(-starts_with("object")) %>% 
+   as.data.frame()
   lag include model nthresh ortho is_same is_same_tssDvec2 is_same_tsD_2ver
1    1    both   VAR      NA  TRUE    TRUE             TRUE             TRUE
2    1    both   VAR      NA FALSE    TRUE             TRUE             TRUE
3    1   const   VAR      NA  TRUE    TRUE             TRUE             TRUE
4    1   const   VAR      NA FALSE    TRUE             TRUE             TRUE
5    1    none   VAR      NA  TRUE    TRUE             TRUE             TRUE
6    1    none   VAR      NA FALSE    TRUE             TRUE             TRUE
7    1   trend   VAR      NA  TRUE   FALSE            FALSE             TRUE
8    1   trend   VAR      NA FALSE   FALSE            FALSE             TRUE
9    2    both   VAR      NA  TRUE    TRUE             TRUE             TRUE
10   2    both   VAR      NA FALSE    TRUE             TRUE             TRUE
11   2   const   VAR      NA  TRUE    TRUE             TRUE             TRUE
12   2   const   VAR      NA FALSE    TRUE             TRUE             TRUE
13   2    none   VAR      NA  TRUE    TRUE             TRUE             TRUE
14   2    none   VAR      NA FALSE    TRUE             TRUE             TRUE
15   2   trend   VAR      NA  TRUE   FALSE            FALSE             TRUE
16   2   trend   VAR      NA FALSE   FALSE            FALSE             TRUE
   comp_irf_tsDOld_tsDNew
1                    TRUE
2                    TRUE
3                    TRUE
4                    TRUE
5                    TRUE
6                    TRUE
7                    TRUE
8                    TRUE
9                    TRUE
10                   TRUE
11                   TRUE
12                   TRUE
13                   TRUE
14                   TRUE
15                   TRUE
16                   TRUE
> 
> ############################
> ### VECM
> ############################
> 
> models_VECM <- models_multivariate %>% 
+   filter(model == "VECM") %>% 
+   mutate(irf = map(object, ~irf_any(.,  boot = TRUE, runs = 1, seed = 7, ortho = FALSE)))
> 
> ## show two first of first componment
> models_VECM %>% 
+   mutate(irf = map(irf, irf_extract_here)) %>% 
+   dplyr::select(-object, -object_vars) %>% 
+   unnest(irf) %>% 
+   as.data.frame() %>% 
+   mutate(across(where(is.numeric), ~round(., 6)))
   lag include model nthresh     type    dolcan   cpiUSA
1    1    both  VECM      NA      irf  1.000000 0.000000
2    1    both  VECM      NA      irf  1.150497 0.536847
3    1    both  VECM      NA Upper_CI  0.000000 1.000000
4    1    both  VECM      NA Upper_CI -0.001956 1.378209
5    1   const  VECM      NA      irf  1.000000 0.000000
6    1   const  VECM      NA      irf  1.158429 0.591890
7    1   const  VECM      NA Upper_CI  0.000000 1.000000
8    1   const  VECM      NA Upper_CI -0.001181 1.388875
9    1    none  VECM      NA      irf  1.000000 0.000000
10   1    none  VECM      NA      irf  1.167443 1.150247
11   1    none  VECM      NA Upper_CI  0.000000 1.000000
12   1    none  VECM      NA Upper_CI  0.005716 1.823742
13   1   trend  VECM      NA      irf  1.000000 0.000000
14   1   trend  VECM      NA      irf  1.162446 0.746819
15   1   trend  VECM      NA Upper_CI  0.000000 1.000000
16   1   trend  VECM      NA Upper_CI -0.000421 1.426727
17   2    both  VECM      NA      irf  1.000000 0.000000
18   2    both  VECM      NA      irf  1.152608 0.652006
19   2    both  VECM      NA Upper_CI  0.000000 1.000000
20   2    both  VECM      NA Upper_CI  0.000536 1.445896
21   2   const  VECM      NA      irf  1.000000 0.000000
22   2   const  VECM      NA      irf  1.162481 0.711727
23   2   const  VECM      NA Upper_CI  0.000000 1.000000
24   2   const  VECM      NA Upper_CI  0.000778 1.443891
25   2    none  VECM      NA      irf  1.000000 0.000000
26   2    none  VECM      NA      irf  1.171534 1.189804
27   2    none  VECM      NA Upper_CI  0.000000 1.000000
28   2    none  VECM      NA Upper_CI  0.001019 1.644230
29   2   trend  VECM      NA      irf  1.000000 0.000000
30   2   trend  VECM      NA      irf  1.167209 0.885486
31   2   trend  VECM      NA Upper_CI  0.000000 1.000000
32   2   trend  VECM      NA Upper_CI  0.000896 1.467909
> 
> 
> ## plot 1
> plot(models_VECM$irf[[1]])
>   
> 
> 
> ############################
> ### TVAR
> ############################
> 
> 
> models_TVAR <- models_multivariate %>% 
+   filter(model == "TVAR")
> 
> ## test 1
> tvar_1 <- models_TVAR$object[[1]]
> 
> irf(tvar_1, runs = 2, seed = 123)

Impulse response coefficients
$dolcan
           dolcan       cpiUSA
 [1,] 0.012915469 -0.003809746
 [2,] 0.012696014  0.005452775
 [3,] 0.012458133  0.014587434
 [4,] 0.012202462  0.023580589
 [5,] 0.011929656  0.032419010
 [6,] 0.011640391  0.041089896
 [7,] 0.011335361  0.049580889
 [8,] 0.011015279  0.057880091
 [9,] 0.010680874  0.065976078
[10,] 0.010332889  0.073857911
[11,] 0.009972079  0.081515151

$cpiUSA
             dolcan    cpiUSA
 [1,]  0.0000000000 0.1603999
 [2,] -0.0003863422 0.1609147
 [3,] -0.0007670853 0.1611538
 [4,] -0.0011416644 0.1611202
 [5,] -0.0015095317 0.1608175
 [6,] -0.0018701580 0.1602497
 [7,] -0.0022230327 0.1594211
 [8,] -0.0025676651 0.1583364
 [9,] -0.0029035841 0.1570008
[10,] -0.0032303396 0.1554197
[11,] -0.0035475025 0.1535988


Lower Band, CI= 0.05 
$cpiUSA
            dolcan     cpiUSA
 [1,]  0.000000000 0.15574683
 [2,] -0.001415551 0.14673544
 [3,] -0.002709885 0.13722747
 [4,] -0.003881594 0.12734160
 [5,] -0.004930428 0.11719020
 [6,] -0.005857201 0.10687889
 [7,] -0.006663698 0.09650616
 [8,] -0.007352591 0.08616311
 [9,] -0.007927336 0.07593325
[10,] -0.008392087 0.06589238
[11,] -0.008751603 0.05610859

$dolcan
           dolcan       cpiUSA
 [1,] 0.013109678 -0.002907084
 [2,] 0.012945142  0.010748397
 [3,] 0.012640610  0.020697951
 [4,] 0.012138876  0.029107229
 [5,] 0.011574632  0.036668717
 [6,] 0.010957403  0.043387383
 [7,] 0.010296409  0.049274807
 [8,] 0.009600513  0.054348553
 [9,] 0.008878174  0.058631533
[10,] 0.008137405  0.062151381
[11,] 0.007385741  0.064939835


Upper Band, CI= 0.05 
$cpiUSA
             dolcan    cpiUSA
 [1,]  0.0000000000 0.1744308
 [2,] -0.0006466191 0.1758129
 [3,] -0.0012860284 0.1765828
 [4,] -0.0019161469 0.1767409
 [5,] -0.0025349603 0.1762896
 [6,] -0.0031405233 0.1752329
 [7,] -0.0037309615 0.1735766
 [8,] -0.0043044738 0.1713282
 [9,] -0.0048593344 0.1684968
[10,] -0.0053938944 0.1650932
[11,] -0.0059065841 0.1611299

$dolcan
           dolcan     cpiUSA
 [1,] 0.013417315 0.00135295
 [2,] 0.013070069 0.01144315
 [3,] 0.012732954 0.02435132
 [4,] 0.012474215 0.03785511
 [5,] 0.012170165 0.05121368
 [6,] 0.011822183 0.06438159
 [7,] 0.011431776 0.07731419
 [8,] 0.011000576 0.08996768
 [9,] 0.010530334 0.10229931
[10,] 0.010022911 0.11426748
[11,] 0.009480272 0.12583181

> 
> ## regime specific for TVAR
> models_TVAR_irf <- models_TVAR  %>% 
+   mutate(irf_L = map(object, ~irf_any(.,  boot = TRUE, runs = 1, seed = 7, ortho = FALSE, regime = "L")))
> 
> ## show two first of first componment
> models_TVAR_irf %>% 
+   mutate(irf = map(irf_L, irf_extract_here)) %>%
+   dplyr::select(-object, -object_vars, -irf_L ) %>% 
+   unnest(irf) %>% 
+   as.data.frame() %>% 
+   mutate(across(where(is.numeric), ~round(., 6)))
   lag include model nthresh     type    dolcan    cpiUSA
1    1    both  TVAR       1      irf  1.000000  0.000000
2    1    both  TVAR       1      irf  0.982298  0.718112
3    1    both  TVAR       1 Upper_CI  0.000000  1.000000
4    1    both  TVAR       1 Upper_CI -0.002896  0.993410
5    1    both  TVAR       2      irf  1.000000  0.000000
6    1    both  TVAR       2      irf  0.982298  0.718112
7    1    both  TVAR       2 Upper_CI  0.000000  1.000000
8    1    both  TVAR       2 Upper_CI -0.002680  0.995241
9    1   const  TVAR       1      irf  1.000000  0.000000
10   1   const  TVAR       1      irf  0.993968  1.341579
11   1   const  TVAR       1 Upper_CI  0.000000  1.000000
12   1   const  TVAR       1 Upper_CI  0.000894  0.995010
13   1   const  TVAR       2      irf  1.000000  0.000000
14   1   const  TVAR       2      irf  1.031097  0.593753
15   1   const  TVAR       2 Upper_CI  0.000000  1.000000
16   1   const  TVAR       2 Upper_CI  0.003019  0.983623
17   1    none  TVAR       1      irf  1.000000  0.000000
18   1    none  TVAR       1      irf  1.001838  0.305646
19   1    none  TVAR       1 Upper_CI  0.000000  1.000000
20   1    none  TVAR       1 Upper_CI  0.000357  1.002424
21   1    none  TVAR       2      irf  1.000000  0.000000
22   1    none  TVAR       2      irf  0.999474  0.306317
23   1    none  TVAR       2 Upper_CI  0.000000  1.000000
24   1    none  TVAR       2 Upper_CI  0.000851  0.999196
25   1   trend  TVAR       1      irf  1.000000  0.000000
26   1   trend  TVAR       1      irf  1.003846  0.685483
27   1   trend  TVAR       1 Upper_CI  0.000000  1.000000
28   1   trend  TVAR       1 Upper_CI  0.002103  0.980364
29   1   trend  TVAR       2      irf  1.000000  0.000000
30   1   trend  TVAR       2      irf  1.028017  0.408567
31   1   trend  TVAR       2 Upper_CI  0.000000  1.000000
32   1   trend  TVAR       2 Upper_CI -0.000175  0.993426
33   2    both  TVAR       1      irf  1.000000  0.000000
34   2    both  TVAR       1      irf  1.116408 -0.057891
35   2    both  TVAR       1 Upper_CI  0.000000  1.000000
36   2    both  TVAR       1 Upper_CI  0.005069  1.179249
37   2    both  TVAR       2      irf  1.000000  0.000000
38   2    both  TVAR       2      irf  1.116408 -0.057891
39   2    both  TVAR       2 Upper_CI  0.000000  1.000000
40   2    both  TVAR       2 Upper_CI  0.006126  1.142099
41   2   const  TVAR       1      irf  1.000000  0.000000
42   2   const  TVAR       1      irf  1.143821  0.835415
43   2   const  TVAR       1 Upper_CI  0.000000  1.000000
44   2   const  TVAR       1 Upper_CI  0.001238  1.366002
45   2   const  TVAR       2      irf  1.000000  0.000000
46   2   const  TVAR       2      irf  1.143821  0.835415
47   2   const  TVAR       2 Upper_CI  0.000000  1.000000
48   2   const  TVAR       2 Upper_CI -0.003818  1.188527
49   2    none  TVAR       1      irf  1.000000  0.000000
50   2    none  TVAR       1      irf  1.145264  0.662497
51   2    none  TVAR       1 Upper_CI  0.000000  1.000000
52   2    none  TVAR       1 Upper_CI  0.001583  1.433116
53   2    none  TVAR       2      irf  1.000000  0.000000
54   2    none  TVAR       2      irf  1.196393 -0.036864
55   2    none  TVAR       2 Upper_CI  0.000000  1.000000
56   2    none  TVAR       2 Upper_CI  0.000725  1.203324
57   2   trend  TVAR       1      irf  1.000000  0.000000
58   2   trend  TVAR       1      irf  1.145049  0.703506
59   2   trend  TVAR       1 Upper_CI  0.000000  1.000000
60   2   trend  TVAR       1 Upper_CI  0.001119  1.396871
61   2   trend  TVAR       2      irf  1.000000  0.000000
62   2   trend  TVAR       2      irf  1.155769  1.053381
63   2   trend  TVAR       2 Upper_CI  0.000000  1.000000
64   2   trend  TVAR       2 Upper_CI -0.000626  1.438063
> 
> 
> ## plot 1
> plot(models_TVAR_irf$irf_L[[1]])
> 
> ############################
> ### TVECM
> ############################
> 
> models_TVECM <- models_multivariate %>% 
+   filter(model == "TVECM")
> 
> ## test 1
> tvecm_1 <- models_TVECM$object[[1]]
> tsDyn:::irf_1(x=tvecm_1 , n.ahead = 10, cumulative = FALSE, regime = "L", ortho = TRUE)
          dolcan        cpiUSA impulse
1   1.259456e-02 -0.0066571462  dolcan
2   1.412961e-02  0.0004987004  dolcan
3   1.357007e-02  0.0117603211  dolcan
4   1.264754e-02  0.0232666549  dolcan
5   1.171809e-02  0.0340318020  dolcan
6   1.085315e-02  0.0438852790  dolcan
7   1.006270e-02  0.0528540463  dolcan
8   9.343524e-03  0.0610059322  dolcan
9   8.689924e-03  0.0684126651  dolcan
10  8.096086e-03  0.0751417442  dolcan
11  7.556583e-03  0.0812550279  dolcan
12  0.000000e+00  0.1510675271  cpiUSA
13 -1.719667e-03  0.1704146708  cpiUSA
14 -2.018136e-03  0.1700704447  cpiUSA
15 -1.823452e-03  0.1666178104  cpiUSA
16 -1.521100e-03  0.1628643752  cpiUSA
17 -1.215673e-03  0.1593240316  cpiUSA
18 -9.309408e-04  0.1560789723  cpiUSA
19 -6.705814e-04  0.1531244399  cpiUSA
20 -4.336612e-04  0.1504388380  cpiUSA
21 -2.183345e-04  0.1479986866  cpiUSA
22 -2.269402e-05  0.1457817803  cpiUSA
> tsDyn:::irf_1(x=tvecm_1 , n.ahead = 10, cumulative = FALSE, regime = "L", ortho = FALSE)
          dolcan    cpiUSA impulse
1   1.0000000000 0.0000000  dolcan
2   1.1158649567 0.6358639  dolcan
3   1.0703936124 1.5288252  dolcan
4   0.9978264483 2.4303404  dolcan
5   0.9250867275 3.2719535  dolcan
6   0.8574797401 4.0419260  dolcan
7   0.7957147322 4.7426864  dolcan
8   0.7395237046 5.3796034  dolcan
9   0.6884572617 5.9582967  dolcan
10  0.6420602816 6.4840435  dolcan
11  0.5999085928 6.9616776  dolcan
12  0.0000000000 1.0000000  cpiUSA
13 -0.0113834300 1.1280695  cpiUSA
14 -0.0133591680 1.1257909  cpiUSA
15 -0.0120704403 1.1029360  cpiUSA
16 -0.0100690100 1.0780899  cpiUSA
17 -0.0080472173 1.0546544  cpiUSA
18 -0.0061624153 1.0331735  cpiUSA
19 -0.0044389510 1.0136159  cpiUSA
20 -0.0028706447 0.9958384  cpiUSA
21 -0.0014452778 0.9796856  cpiUSA
22 -0.0001502243 0.9650107  cpiUSA
> irf(x=tvecm_1, runs = 2, seed = 123)

Impulse response coefficients
$dolcan
           dolcan        cpiUSA
 [1,] 0.012594557 -0.0066571462
 [2,] 0.014129606  0.0004987004
 [3,] 0.013570067  0.0117603211
 [4,] 0.012647537  0.0232666549
 [5,] 0.011718088  0.0340318020
 [6,] 0.010853149  0.0438852790
 [7,] 0.010062699  0.0528540463
 [8,] 0.009343524  0.0610059322
 [9,] 0.008689924  0.0684126651
[10,] 0.008096086  0.0751417442
[11,] 0.007556583  0.0812550279

$cpiUSA
             dolcan    cpiUSA
 [1,]  0.000000e+00 0.1510675
 [2,] -1.719667e-03 0.1704147
 [3,] -2.018136e-03 0.1700704
 [4,] -1.823452e-03 0.1666178
 [5,] -1.521100e-03 0.1628644
 [6,] -1.215673e-03 0.1593240
 [7,] -9.309408e-04 0.1560790
 [8,] -6.705814e-04 0.1531244
 [9,] -4.336612e-04 0.1504388
[10,] -2.183345e-04 0.1479987
[11,] -2.269402e-05 0.1457818


Lower Band, CI= 0.05 
$cpiUSA
             dolcan    cpiUSA
 [1,]  0.0000000000 0.1518679
 [2,] -0.0036766033 0.1329288
 [3,] -0.0037303415 0.1320815
 [4,] -0.0035104769 0.1372203
 [5,] -0.0030529103 0.1408389
 [6,] -0.0025695700 0.1439257
 [7,] -0.0021164219 0.1468103
 [8,] -0.0017035816 0.1494577
 [9,] -0.0013298109 0.1518770
[10,] -0.0009918396 0.1526318
[11,] -0.0006863122 0.1503966

$dolcan
           dolcan        cpiUSA
 [1,] 0.012761098 -0.0122545261
 [2,] 0.013166774  0.0001548694
 [3,] 0.011317071  0.0094704856
 [4,] 0.010186703  0.0001172342
 [5,] 0.009204399 -0.0071924029
 [6,] 0.008275847 -0.0139656825
 [7,] 0.007425881 -0.0202174802
 [8,] 0.006650625 -0.0259316579
 [9,] 0.005942264 -0.0311576357
[10,] 0.005294965 -0.0359393743
[11,] 0.004703522 -0.0403143296


Upper Band, CI= 0.05 
$cpiUSA
             dolcan    cpiUSA
 [1,]  0.000000e+00 0.1593005
 [2,] -3.018230e-03 0.1785313
 [3,] -2.967110e-03 0.1765646
 [4,] -2.293506e-03 0.1722362
 [5,] -1.842602e-03 0.1680351
 [6,] -1.417866e-03 0.1642652
 [7,] -1.021275e-03 0.1608892
 [8,] -6.595127e-04 0.1578498
 [9,] -3.292642e-04 0.1551075
[10,] -2.747135e-05 0.1540905
[11,]  2.482834e-04 0.1561159

$dolcan
           dolcan      cpiUSA
 [1,] 0.012986537 0.001034214
 [2,] 0.015124778 0.020232192
 [3,] 0.014625889 0.011250617
 [4,] 0.013569122 0.020816218
 [5,] 0.012474651 0.029261791
 [6,] 0.011455171 0.036801019
 [7,] 0.010528433 0.043584021
 [8,] 0.009689996 0.049703927
 [9,] 0.008931985 0.055228841
[10,] 0.008246748 0.060217201
[11,] 0.007627298 0.064721239

> 
> ## regime specific for TVECM
> models_TVECM_irf <- models_TVECM   %>% 
+   mutate(irf_L = map(object, ~suppressWarnings(irf_any(.,  boot = TRUE, runs = 1, seed = 7, ortho = FALSE, regime = "L"))))
> 
> ## show two first of first componment
> models_TVECM_irf %>% 
+   mutate(irf = map(irf_L, irf_extract_here)) %>% 
+   select(-object, -object_vars, -irf_L ) %>% 
+   unnest(irf) %>% 
+   as.data.frame() %>% 
+   mutate(across(where(is.numeric), ~round(., 6)))
   lag include model nthresh     type    dolcan   cpiUSA
1    1    both TVECM       1      irf  1.000000 0.000000
2    1    both TVECM       1      irf  1.115865 0.635864
3    1    both TVECM       1 Upper_CI  0.000000 1.000000
4    1    both TVECM       1 Upper_CI  0.001099 0.992494
5    1    both TVECM       2      irf  1.000000 0.000000
6    1    both TVECM       2      irf  1.115865 0.635864
7    1    both TVECM       2 Upper_CI  0.000000 1.000000
8    1    both TVECM       2 Upper_CI -0.002071 0.795772
9    1   const TVECM       1      irf  1.000000 0.000000
10   1   const TVECM       1      irf  1.087586 0.899218
11   1   const TVECM       1 Upper_CI  0.000000 1.000000
12   1   const TVECM       1 Upper_CI  0.001768 0.899371
13   1   const TVECM       2      irf  1.000000 0.000000
14   1   const TVECM       2      irf  1.087586 0.899218
15   1   const TVECM       2 Upper_CI  0.000000 1.000000
16   1   const TVECM       2 Upper_CI -0.003737 0.945145
17   1    none TVECM       1      irf  1.000000 0.000000
18   1    none TVECM       1      irf  1.095626 0.465547
19   1    none TVECM       1 Upper_CI  0.000000 1.000000
20   1    none TVECM       1 Upper_CI  0.005645 1.653400
21   1    none TVECM       2      irf  1.000000 0.000000
22   1    none TVECM       2      irf  1.142178 1.809654
23   1    none TVECM       2 Upper_CI  0.000000 1.000000
24   1    none TVECM       2 Upper_CI  0.005597 1.745848
25   1   trend TVECM       1      irf  1.000000 0.000000
26   1   trend TVECM       1      irf  1.183445 0.872582
27   1   trend TVECM       1 Upper_CI  0.000000 1.000000
28   1   trend TVECM       1 Upper_CI  0.000343 1.610538
29   1   trend TVECM       2      irf  1.000000 0.000000
30   1   trend TVECM       2      irf  1.078900 1.489184
31   1   trend TVECM       2 Upper_CI  0.000000 1.000000
32   1   trend TVECM       2 Upper_CI -0.002040 0.820806
33   2    both TVECM       1      irf  1.000000 0.000000
34   2    both TVECM       1      irf  1.168665 0.281654
35   2    both TVECM       1 Upper_CI  0.000000 1.000000
36   2    both TVECM       1 Upper_CI -0.001356 1.279373
37   2    both TVECM       2      irf  1.000000 0.000000
38   2    both TVECM       2      irf  1.139679 0.236860
39   2    both TVECM       2 Upper_CI  0.000000 1.000000
40   2    both TVECM       2 Upper_CI -0.008496 1.213897
41   2   const TVECM       1      irf  1.000000 0.000000
42   2   const TVECM       1      irf  1.179502 0.472377
43   2   const TVECM       1 Upper_CI  0.000000 1.000000
44   2   const TVECM       1 Upper_CI  0.004413 1.310172
45   2   const TVECM       2      irf  1.000000 0.000000
46   2   const TVECM       2      irf  1.112088 0.587852
47   2   const TVECM       2 Upper_CI  0.000000 1.000000
48   2   const TVECM       2 Upper_CI -0.008211 1.055222
49   2    none TVECM       1      irf  1.000000 0.000000
50   2    none TVECM       1      irf  1.121259 0.092243
51   2    none TVECM       1 Upper_CI  0.000000 1.000000
52   2    none TVECM       1 Upper_CI  0.010255 1.583673
53   2    none TVECM       2      irf  1.000000 0.000000
54   2    none TVECM       2      irf  1.231963 0.074279
55   2    none TVECM       2 Upper_CI  0.000000 1.000000
56   2    none TVECM       2 Upper_CI  0.015478 1.536769
57   2   trend TVECM       1      irf  1.000000 0.000000
58   2   trend TVECM       1      irf  1.184091 0.748835
59   2   trend TVECM       1 Upper_CI  0.000000 1.000000
60   2   trend TVECM       1 Upper_CI  0.001301 1.348732
61   2   trend TVECM       2      irf  1.000000 0.000000
62   2   trend TVECM       2      irf  1.158615 0.715514
63   2   trend TVECM       2 Upper_CI  0.000000 1.000000
64   2   trend TVECM       2 Upper_CI -0.007753 1.246237
> 
> 
> ## plot 1
> plot(models_TVECM_irf$irf_L[[1]])
> 
> proc.time()
   user  system elapsed 
 42.242   0.322  43.696