search

yahoo / egads

A Java package to automatically detect anomalies in large scale time-series data
Other
1.17k stars 331 forks source link

Using an "expected" time-series for forecasting #40

Open gohar94 opened 7 years ago

gohar94 commented 7 years ago

Hi,

I am trying to specify a list of time-series that are my "normal" behavior and a list or an individual time-series which I want to check for anomalies against the "normal" list - is there a way to do this under the current release? If not, is it preferable that I work on adding another TSMM model which takes in a list of .csv files as "normal" time-series and averages them at each time unit to calculate an "expected" time-series and then compares the other time-series with this?

Thanks.