将数据帧转换为 r 中的时间序列

Question

I am currently trying to convert a data.frame to a time series. The data frame looks like this:

All I want to do is be able to plot the doc data as a function of time and run a statistical test on it.

Any help would be greatly appreciated!

This is what my code currently looks like:

x=aggregate( doc ~ mo + yr , B , mean )
x$Date <- as.yearmon(paste(x$yr, x$mo), "%Y %m")
df_ts <- xts(x, order.by = x$Date)
keeps <- "doc"
df_ts <- df_ts[ , keeps, drop = FALSE]
df_ts_1 <- as.ts(df_ts, start = head(index(df_ts), 1), end = 
tail(index(df_ts), 1))

The issue I'm running into is that the months and years are not in sequential order so when I try to apply a as.tf function, the data does not fill in correctly.

Answer 1

Using DF defined reproducibly in the Note at the end read the data frame into a zoo object converting the Date column to yearmon class and plot using ggplot2. If there exist duplicate dates (there are none in the example data) then add the aggregate = mean argument to read.zoo .

library(ggplot2)
library(zoo)

z <- read.zoo(DF[c("Date", "doc")], FUN = as.yearmon, format = "%b %Y")
autoplot(z) + scale_x_yearmon()

This would also work:

tt <- as.ts(z)
plot(na.approx(tt), ylab = "tt")

Note

In the future please do not use images. I have retyped the first three rows this time.

DF <- data.frame(month = c("02", "10", "12"), year = c(1998, 2000, 2000),
 doc = c(1.55, 2.2, 0.96), Date = c("Feb 1998", "Oct 2000", "Dec 2000"),
 stringsAsFactors = FALSE)