R中每小时数据的时间序列预测

Question

I have a CSV that has close to 2 months of data with data captured on an hourly basis.

The data set is as follows

Date_Time                    Freq
2018-06-01 01:00:00           232
2018-06-01 02:00:00           99
2018-06-01 03:00:00           90

Most time series examples mention this on yearly data. I am a bit lost with how to do forecasting on an hourly basis. What would be the best way to do it in R. There are many examples online that uses h2o, knn, ARIMA etc.

Answer 1

 # R
df <- read.csv('../examples/example_yosemite_temps.csv')
m <- prophet(df, changepoint.prior.scale=0.01)
future <- make_future_dataframe(m, periods = 300, freq = 60 * 60)
fcst <- predict(m, future)
plot(m, fcst)

https://facebook.github.io/prophet/docs/non-daily_data.html