如何使用寓言r将预测加在一起？

如何解决如何使用寓言r将预测加在一起？

我正在尝试将每个地区和每个季度的预测加在一起。我试图总结一个特定的变量，但这没有用。我也尝试过按区域分组，但我相信要使此方法有效，您需要以某种方式进行汇总。我正在尝试将两个预测加在一起，同时也能够使用precision（）来查看准确性的变化。到目前为止，我还没有真正找到一种好的方法。

tourism %>% 
  filter(Quarter >= yearquarter("2008 Q1")) %>%
  filter(Purpose == "Holiday" & State == "New South Wales") %>%
  filter(Region %in% c("North Coast NSW","South Coast","Sydney")) %>%
  mutate(Demand = case_when(
    Region == "Sydney" ~ 0.03*Trips*5,Region == "South Coast"~0.04*Trips*2,Region == "North Coast NSW" ~ 0.04 *Trips*2
  )) -> D

DTR <- D %>% filter(Quarter <= yearquarter("2016 Q4"))
DTE <- D %>% filter(Quarter >= yearquarter("2017 Q1"))


m <- DTR %>% 
  model(m.auto = ETS(Demand),m.AAM = ETS(Demand ~ error("A") + trend("A") + season("M")),m.AAdM = ETS(Demand ~ error("A") + trend("Ad") + season("M")))


m  %>%
  glance()

m %>% 
  select(m.auto) %>%
  glance()
m.auto <- m %>%
  select(m.auto)
  

f <- m %>% 
  select(m.auto) %>%
  forecast(h = 4)

rbind(m.auto %>% accuracy(),f %>% accuracy(data = D))

以下是预测表：

structure(list(Region = c("North Coast NSW","North Coast NSW","Sydney","Sydney"
),State = c("New South Wales","New South Wales","New South Wales"),Purpose = c("Holiday","Holiday","Holiday"),.model = c("m.auto","m.auto","m.auto"),Quarter = structure(c(17167,17257,17348,17440,17167,17440),fiscal_start = 1,class = c("yearquarter","vctrs_vctr"
)),Demand = structure(list(structure(list(mu = 64.5267676142378,sigma = 7.45649497452699),class = c("dist_normal","dist_default"
)),structure(list(mu = 45.7869538723355,sigma = 5.39507245782213),"dist_default")),structure(list(mu = 42.4573776516376,sigma = 5.09748859911909),structure(list(mu = 53.9721361044215,sigma = 6.59826306707935),structure(list(mu = 61.7503906560609,sigma = 6.75515872707897),structure(list(mu = 35.6173573447291,sigma = 3.97232150452371),structure(list(mu = 27.1314815258956,sigma = 3.08272534449931),structure(list(mu = 38.1701802125742,sigma = 4.41547997477935),structure(list(mu = 95.1714054395916,sigma = 8.18541561715263),structure(list(mu = 85.2330580025708,sigma = 8.18541582507749),structure(list(mu = 79.7805187916752,sigma = 8.18541603300235),structure(list(mu = 77.9011650250759,sigma = 8.18541646640782),"dist_default"))),vars = "Demand",class = c("distribution","vctrs_vctr","list")),.mean = c(64.5267676142378,45.7869538723355,42.4573776516376,53.9721361044215,61.7503906560609,35.6173573447291,27.1314815258956,38.1701802125742,95.1714054395916,85.2330580025708,79.7805187916752,77.9011650250759)),row.names = c(NA,-12L),key = structure(list(
    Region = c("North Coast NSW","Sydney"),"m.auto"
    ),.rows = structure(list(1:4,5:8,9:12),ptype = integer(0),class = c("vctrs_list_of","list"))),3L),class = c("tbl_df","tbl","data.frame"),.drop = TRUE),index = structure("Quarter",ordered = TRUE),index2 = "Quarter",interval = structure(list(
    year = 0,quarter = 1,month = 0,week = 0,day = 0,hour = 0,minute = 0,second = 0,millisecond = 0,microsecond = 0,nanosecond = 0,unit = 0),.regular = TRUE,class = c("interval","vctrs_rcrd","vctrs_vctr")),response = "Demand",dist = "Demand",model_cn = ".model",class = c("fbl_ts","tbl_ts","tbl_df","data.frame"))

如果您需要更多信息，请告诉我。