CODEWITHSUNDEEP
pythontestingtime-seriespartialautocorrelation
John hall
John hall

Reputation: 1

Seasonality and period of a non stationary time series

So I have a non stationary time series that is hourly, daily and monthly recorded for a year, like this:

Hours
2020-01-01 00:00:00     4.0
2020-01-01 01:00:00     3.0
2020-01-01 02:00:00     4.0
2020-01-01 03:00:00     4.0
2020-01-01 04:00:00    14.0
                       ... 
2020-12-31 19:00:00    60.0
2020-12-31 20:00:00    59.0
2020-12-31 21:00:00    72.0
2020-12-31 22:00:00    70.0
2020-12-31 23:00:00    68.0

And I have the ACF and PACF plots for the serie.

ACF

and

PACF

Looking at the acf test I can see big pikes at lags=24,48,72. Does that mean that I have a period of 24 hours, or a seasonality of 24 h?

Also, I applied the the Ljung-Box test and for every lag I obtain a p value of 0, which means that my values are showing dependence on each other. I also don´t understand what is the meaning of this test in terms of seasonality.

With this done I need to know what is the period and seasonality of this time serie but I am confused. Need help.

Thank you

Upvotes: 0

Views: 103

Answers (0)

Related Questions