Beyond GARCH 2 • ahead

Sequel of Beyond GARCH vignette, but using Statistical models for modeling the volatility.

# Default model for volatility (Ridge regression for volatility)
(obj_ridge <- ahead::mlarchf(fpp2::goog200, h=20L, B=500L, ml=FALSE, stat_model=forecast::thetaf))

##     Point Forecast    Lo 95    Hi 95
## 201       535.1202 505.3340 574.0000
## 202       534.6643 508.3409 571.1790
## 203       536.6966 517.2748 575.5046
## 204       537.0459 492.9445 592.3481
## 205       537.3298 510.5214 576.1656
## 206       538.1587 502.7001 560.6767
## 207       538.6766 516.8262 571.6381
## 208       534.7889 495.9735 569.4959
## 209       538.4172 498.5907 592.0333
## 210       540.6477 509.1067 565.6255
## 211       539.2109 498.9948 574.0345
## 212       541.0359 515.8817 565.7521
## 213       543.3602 522.8714 586.9719
## 214       538.7158 499.0550 568.5013
## 215       542.8687 501.2013 568.7869
## 216       543.9402 503.6002 576.4102
## 217       543.2257 516.7300 565.1119
## 218       546.2282 507.6338 585.7978
## 219       544.4680 512.0647 564.0864
## 220       547.5491 521.3138 587.5447

plot(obj_ridge)

(obj_ridge <- ahead::mlarchf(fpp2::goog200, h=20L, B=500L, ml=FALSE, stat_model=forecast::meanf))

##     Point Forecast    Lo 95    Hi 95
## 201       532.6337 480.3920 562.5208
## 202       535.3076 500.7168 576.2883
## 203       535.8617 492.6612 580.0431
## 204       536.7824 505.4796 570.1889
## 205       537.4004 503.6281 567.1039
## 206       537.3192 492.1927 584.2525
## 207       539.4490 502.0956 575.6489
## 208       538.9314 510.0769 564.4209
## 209       537.0118 496.2940 570.1630
## 210       541.0988 513.7256 571.8766
## 211       540.1650 509.2756 571.6424
## 212       541.5303 505.9652 569.0016
## 213       541.7569 518.9127 581.4046
## 214       543.2300 502.8549 565.8640
## 215       543.8087 519.9057 570.4386
## 216       540.6660 481.1621 577.8674
## 217       544.1822 504.2330 572.7532
## 218       543.7387 510.7820 586.3453
## 219       546.1519 518.3156 580.1015
## 220       546.8818 508.6304 587.1924

plot(obj_ridge)

(obj_ridge <- ahead::mlarchf(fpp2::goog200, h=20L, B=500L, ml=FALSE, stat_model=forecast::auto.arima))

##     Point Forecast    Lo 95    Hi 95
## 201       532.9034 493.0183 553.1978
## 202       534.3819 481.2459 575.7542
## 203       537.1642 514.9170 585.5694
## 204       538.0364 512.3756 565.8655
## 205       534.6230 511.9038 572.2574
## 206       537.5614 509.7811 568.3510
## 207       539.1844 512.7942 576.2216
## 208       539.5243 509.3811 591.5045
## 209       539.7391 505.9189 572.9482
## 210       539.5957 499.7606 567.7732
## 211       540.0432 509.8230 567.0591
## 212       543.0360 507.8170 589.1519
## 213       540.1045 508.9871 575.7777
## 214       542.5956 525.1924 604.0171
## 215       543.7266 512.1619 575.8865
## 216       539.5808 493.5483 571.4434
## 217       545.2709 519.2288 570.3752
## 218       541.0958 504.6950 587.2074
## 219       546.1951 517.9425 583.8326
## 220       546.1588 516.4791 589.6906

plot(obj_ridge)

(obj_ridge <- ahead::mlarchf(fpp2::goog200, h=20L, B=500L, ml=FALSE, stat_model=forecast::ets))

##     Point Forecast    Lo 95    Hi 95
## 201       532.5195 489.4041 566.2702
## 202       534.1427 508.6353 568.5833
## 203       534.4201 508.9994 570.7575
## 204       536.0049 509.6585 557.9329
## 205       535.0316 503.0271 557.1738
## 206       536.0877 509.8492 584.5520
## 207       537.7928 508.7091 571.2139
## 208       538.0152 508.3012 571.2743
## 209       539.0516 468.5178 578.6762
## 210       540.6423 504.7203 578.3843
## 211       540.6572 508.0521 567.5746
## 212       540.4657 499.3215 575.6383
## 213       534.8167 486.1280 579.3049
## 214       544.3092 518.7348 581.8536
## 215       542.1076 515.2904 581.3185
## 216       543.8792 511.7354 595.5889
## 217       542.7291 495.7240 579.1987
## 218       545.5925 506.6893 578.5368
## 219       546.9548 524.2050 577.4921
## 220       546.2788 504.3733 598.1818

plot(obj_ridge)