How to detect the anomaly in a time series data(specifically) with trend and seasonality present in it?

52
July 16, 2019, at 12:10 PM

I want to detect the outliers in a "time series data" which contains the trend and seasonality components. I want to leave out the peaks which are seasonal and only consider only the other peaks and label them as outliers. As I am new to time series analysis, Please assist me to approach this time series problem.

The coding platform is using is Python.

So far I have tried the ARIMA model. I have trained my model and forecasted for the test data. Then being able to compute the difference between forecasted results with my actual values of test data then able to find out the outliers based on the variance observed.

But I can detect the outliers only in test data. Actually, I have to detect the outliers for the whole time series data I am having.

What is the best approach to do this?

READ ALSO
killing py4j request and java processes associated with it

killing py4j request and java processes associated with it

I am calling java function from python using py4j java gateway

57
Python Threads: event.set does not activate the thread in Task/Job management

Python Threads: event.set does not activate the thread in Task/Job management

I have a Task Broker who uses an internal message queue (redis) where Workers poll tasks fromThe tasks have a timeout and should be finished by a Worker before they are expired

73
How to customize AOSP keyboard?

How to customize AOSP keyboard?

I can try to add tab layout in AOSP keybord but tab click doesn't work and also whatsapp edittext hide behind tab so please help me how to add tab functionality in AOSP keyboard ?

56
Showing error while routing from one screen to another in flutter

Showing error while routing from one screen to another in flutter

I'm moving from one screen to another screen but it shows error in routingMy First screen named as VenueOption and the second one is PlayerOption

66