-
Notifications
You must be signed in to change notification settings - Fork 9
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Default threshold_outlier value for EWMA monitoring class #4
Comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
While experimenting with simulated time-series, I found out that the default value for the
threshold_outlier
argument should be significantly higher (10 could be a more reasonable value, though that would depend on the range of the monitored values too I guess). Whenthreshold_outlier
is too low, all observations following a disturbance are considered extreme outliers and the process value is no longer updated, resulting in the break never being detected.default
sensitivity
also appears to be a bit low (meaning too sensitive). I obtain best performances with a sensitivity of 4; though the synthetic data may not be realistic enough to be fully conclusive on that aspect.The text was updated successfully, but these errors were encountered: