微信公众号搜"智元新知"关注
微信扫一扫可直接关注哦!

R Forecast 'ts' 对象必须有一个或多个观察

如何解决R Forecast 'ts' 对象必须有一个或多个观察

不明白为什么会出现以下错误。我不习惯在 r 中使用 ts 对象。我可以看到数据中有观察。

head(bike_sales)
library(tidyverse)
library(lubridate)
library(zoo)
library(sweep)
library(tidyr)
library(forecast)
library(forecastHybrid)
library(timetk)
data <- bike_sales %>%
  mutate(order.month = as_date(as.yearmon(order.date))) %>%
  group_by(category.secondary,customer.id,order.month) %>%
  summarise(total.qty = sum(quantity))

nested_monthly <- data %>%
  group_by(category.secondary,customer.id) %>%
  rename()
  nest(.key = "series") %>%
  ungroup()

nested_monthly %>%
  mutate(train.ts = map(series,~.x %>% 
                          filter(order.month < make_date(2014,1,1)) %>%
                          tk_ts(frequency = 12,start = c(2011,1),end = c(2013,12),silent = T)),test.ts = map(series,~.x %>% 
                         filter(order.month >= make_date(2014,1)) %>%
                         tk_ts(frequency = 12,start = c(2014,end = c(2015,# aggregate results for yearly predictions
         train.ts.agg = map(train.ts,~.x %>% 
                              stats::filter(.,rep(1,sides = 1) %>%
                              stats::window(start = c(2011,12))),test.ts.agg = map(test.ts,~.x %>% 
                             stats::filter(.,sides = 1) %>%
                             stats::window(start = c(2014,12))))

版权声明:本文内容由互联网用户自发贡献,该文观点与技术仅代表作者本人。本站仅提供信息存储空间服务,不拥有所有权,不承担相关法律责任。如发现本站有涉嫌侵权/违法违规的内容, 请发送邮件至 dio@foxmail.com 举报,一经查实,本站将立刻删除。