主页 / coding / 问题 / 79137706
Accepted
user430997
Asked: 2024-10-29 22:41:39 +0800 CST

如何在 R 中绑定时间序列？

  • 772

这是我正在处理的问题。

我从 R 中的时间序列模型获得了以下两个输出：

library(forecast)
library(dplyr)
set.seed(123)

dates <- seq(as.Date("2000-01-01"), by = "month", length.out = 100)
##################


values1 <- rnorm(100, mean = 100, sd = 100)
ts_data1 <- ts(values1, start = c(2000, 1), frequency = 12)
model1 <- auto.arima(ts_data1)

forecast_data1 <- forecast(model1, h = 12)
forecast_data1$class = 1

####################


values2 <- rnorm(100, mean = 200, sd = 100)
ts_data2 <- ts(values2, start = c(2000, 1), frequency = 12)
model2 <- auto.arima(ts_data2)

forecast_data2 <- forecast(model2, h = 12)
forecast_data2$class = 2

我正在尝试将这两个结合在一起，例如

 rbind(data.frame(forecast_data1), data.frame(forecast_data2))
          Point.Forecast        Lo.80    Hi.80       Lo.95    Hi.95
May 2008       116.88772   2.03451679 231.7409  -58.765094 292.5405
Jun 2008        84.77524 -30.07795384 199.6284  -90.877564 260.4281
Jul 2008        88.15699 -26.69621184 203.0102  -87.495822 263.8098
Aug 2008        97.84920 -17.00400178 212.7024  -77.803612 273.5020
Sep 2008       104.59216 -10.26103639 219.4454  -71.060647 280.2450
Oct 2008       123.46321   8.61001633 238.3164  -52.189594 299.1160
Nov 2008        80.16232 -34.69087853 195.0155  -95.490489 255.8151
Dec 2008       122.86121   8.00801357 237.7144  -52.791597 298.5140
Jan 2009        62.16134 -52.69186110 177.0145 -113.491472 237.8141
Feb 2009        76.41793 -38.43527070 191.2711  -99.234881 252.0707
Mar 2009       114.92293   0.06973034 229.7761  -60.729880 290.5757
Apr 2009       132.14088  17.28768324 246.9941  -43.511927 307.7937
May 20081      173.02664  50.54760673 295.5057  -14.288876 360.3422
Jun 20081      196.17783  73.69880321 318.6569    8.862321 383.4933
Jul 20081      181.38489  58.90585584 303.8639   -5.930626 368.7004
Aug 20081      196.21547  73.73644398 318.6945    8.899962 383.5310
Sep 20081      189.46162  66.98258965 311.9407    2.146107 376.7771
Oct 20081      200.14378  77.66475146 322.6228   12.828269 387.4593
Nov 20081      214.71057  92.23154347 337.1896   27.395061 402.0261
Dec 20081      151.90221  29.42317702 274.3812  -35.413305 339.2177
Jan 20091      176.64285  54.16381860 299.1219  -10.672664 363.9584
Feb 20091      211.48637  89.00734228 333.9654   24.170860 398.8019
Mar 20091      196.08048  73.60144798 318.5595    8.764966 383.3960
Apr 20091      210.85820  88.37916960 333.3372   23.542687 398.1737

但由于某种原因，类变量（即 1 或 2）没有显示出来以表明预测来自哪个模型。

我该如何修复此问题？

2 个回答

  1. 当您将预测对象转换为数据框时，class插槽会丢失。您应该class在转换后添加插槽。例如

    > forecast_data2 <- data.frame(forecast(model2, h = 12))
> 
> forecast_data2$class = 2

> > forecast_data2
         Point.Forecast    Lo.80    Hi.80      Lo.95    Hi.95 class
May 2008       173.0266 50.54761 295.5057 -14.288876 360.3422     2
Jun 2008       196.1778 73.69880 318.6569   8.862321 383.4933     2
Jul 2008       181.3849 58.90586 303.8639  -5.930626 368.7004     2
Aug 2008       196.2155 73.73644 318.6945   8.899962 383.5310     2
Sep 2008       189.4616 66.98259 311.9407   2.146107 376.7771     2
Oct 2008       200.1438 77.66475 322.6228  12.828269 387.4593     2
Nov 2008       214.7106 92.23154 337.1896  27.395061 402.0261     2
Dec 2008       151.9022 29.42318 274.3812 -35.413305 339.2177     2
Jan 2009       176.6428 54.16382 299.1219 -10.672664 363.9584     2
Feb 2009       211.4864 89.00734 333.9654  24.170860 398.8019     2
Mar 2009       196.0805 73.60145 318.5595   8.764966 383.3960     2
Apr 2009       210.8582 88.37917 333.3372  23.542687 398.1737     2
    • 1
  2. Best Answer

    forecast在添加列之前将对象转换为 data.frame class，例如：

    dates <- seq(as.Date("2000-01-01"), by = "month", length.out = 100)
values1 <- rnorm(100, mean = 100, sd = 100)
ts_data1 <- ts(values1, start = c(2000, 1), frequency = 12)
model1 <- auto.arima(ts_data1)

forecast_data1 <- forecast(model1, h = 12)
forecast_data1$class = 1

class(forecast_data1)
#> [1] "forecast"

forecast_data1 <- forecast_data1 |>
  as.data.frame()

forecast_data1$class = 1

    然后第二个也一样：

    forecast_data2 <- forecast(model2, h = 12)

forecast_data2 <- forecast_data2 |>
  as.data.frame()
forecast_data2$class = 2

rbind(forecast_data1, forecast_data2)
#>           Point Forecast        Lo 80    Hi 80       Lo 95    Hi 95 class
#> May 2008       116.88772   2.03451679 231.7409  -58.765094 292.5405     1
#> Jun 2008        84.77524 -30.07795384 199.6284  -90.877564 260.4281     1
#> Jul 2008        88.15699 -26.69621184 203.0102  -87.495822 263.8098     1
#> Aug 2008        97.84920 -17.00400178 212.7024  -77.803612 273.5020     1
#> Sep 2008       104.59216 -10.26103639 219.4454  -71.060647 280.2450     1
#> Oct 2008       123.46321   8.61001633 238.3164  -52.189594 299.1160     1
#> Nov 2008        80.16232 -34.69087853 195.0155  -95.490489 255.8151     1
#> Dec 2008       122.86121   8.00801357 237.7144  -52.791597 298.5140     1
#> Jan 2009        62.16134 -52.69186110 177.0145 -113.491472 237.8141     1
#> Feb 2009        76.41793 -38.43527070 191.2711  -99.234881 252.0707     1
#> Mar 2009       114.92293   0.06973034 229.7761  -60.729880 290.5757     1
#> Apr 2009       132.14088  17.28768324 246.9941  -43.511927 307.7937     1
#> May 20081      173.02664  50.54760673 295.5057  -14.288876 360.3422     2
#> Jun 20081      196.17783  73.69880321 318.6569    8.862321 383.4933     2
#> Jul 20081      181.38489  58.90585584 303.8639   -5.930626 368.7004     2
#> Aug 20081      196.21547  73.73644398 318.6945    8.899962 383.5310     2
#> Sep 20081      189.46162  66.98258965 311.9407    2.146107 376.7771     2
#> Oct 20081      200.14378  77.66475146 322.6228   12.828269 387.4593     2
#> Nov 20081      214.71057  92.23154347 337.1896   27.395061 402.0261     2
#> Dec 20081      151.90221  29.42317702 274.3812  -35.413305 339.2177     2
#> Jan 20091      176.64285  54.16381860 299.1219  -10.672664 363.9584     2
#> Feb 20091      211.48637  89.00734228 333.9654   24.170860 398.8019     2
#> Mar 20091      196.08048  73.60144798 318.5595    8.764966 383.3960     2
#> Apr 20091      210.85820  88.37916960 333.3372   23.542687 398.1737     2

    创建于 2024-10-29，使用reprex v2.1.0

    • 1

相关问题