R建模中的ARIMA？

Question

So im quite new to R, and I need some help!

This is what I've done so far

library("readxl")

my_data <- read_excel(file.choose())

View(my_data)

library(ggplot2)

theme_set(theme_light())

ggplot(my_data, aes(x=DATES)) +
geom_line(aes(y=GDPBD))

ggplot(my_data, aes(x=DATES)) + 
geom_line(aes(y=GDPBD), color = "#00AFBB") +
labs(y="GDP", x="Year")

Basically I'm trying to predict GDP for the upcoming 5 years.

Since i don't know the exact ARIMA (p,d,q) model, i wanted to try the auto.arima function.

Yet it says:

Error in auto.arima(my_data) : auto.arima can only handle univariate time series.

Dates are already yearly.

Can you help me?

Also, do you have other methods that allow me to find d in (p,d,q) using the lags? i dont know when to stop just by looking at the graphs.

EDIT: i know that the process is not stationary

Your help is much appreciated!

Thank you !

Answer 1

convert your data into ts object, like below example

 my_data = ts(my_data)
 forecast = auto.arima(mydata)

Hope answer to your question.