I have data of a daily time series which consists of the date and the sum about all workin hours. So I have one column with dates and another one with numbers. Because it deals about working hours, they are nearly 0 on weekend and on holiday days.
My aim is to predict the workloud of the following weeks or month.
I already tried to implement the answers of https://stackoverflow.com/questions/33128865/starting-a-daily-time-series-in-r, but the output of R consists of very similar values which doesn't take the low values on weekends into account.
Here is a Screenshot of the Time Series:
Why does the auto.arima() don't give an useful output?
So can you help me to get a forecast which takes for example the low values on weekend and on holidays into account?
Thanks for any help!
