如何用R做iterated forecast

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

求大神指导一下，怎么用R写iterated two-quarter ahead out of sample forecast，多谢多谢。

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏0 回帖

关键词：iterated Forecast forecas ERAT Rate 如何

呃，都贴进来吧，都用得上...

One-step forecasts without re-estimation

train <- window(hsales,end=1989.99)
fit <- auto.arima(train)
refit <- Arima(hsales, model=fit)
fc <- window(fitted(refit), start=1990)

Multi-step forecasts without re-estimation

h <- 5
train <- window(hsales,end=1989.99)
test <- window(hsales,start=1990)
n <- length(test) - h + 1
fit <- auto.arima(train)
fc <- ts(numeric(n), start=1990+(h-1)/12, freq=12)
for(i in 1:n)
{  
  x <- window(hsales, end=1989.99 + (i-1)/12)
  refit <- Arima(x, model=fit)
  fc <- forecast(refit, h=h)$mean[h]
}

Multi-step forecasts with re-estimation

h <- 5
train <- window(hsales,end=1989.99)
test <- window(hsales,start=1990)
n <- length(test) - h + 1
fit <- auto.arima(train)
order <- arimaorder(fit)
fcmat <- matrix(0, nrow=n, ncol=h)
for(i in 1:n)
{  
  x <- window(hsales, end=1989.99 + (i-1)/12)
  refit <- Arima(x, order=order[1:3], seasonal=order[4:6])
  fcmat[i,] <- forecast(refit, h=h)$mean
}
A variation on this also re-selects the model at each iteration. Then the second line in the loop is replaced with
refit <- auto.arima(x)

厉害 学习学习