Apparent error in Stepp completeness graph

[Edwar Q S]

When I apply the stepp 1972 completeness analysis algorithm, I get an anomalous graph for the last magnitude window. As shown in the image, apparently there is data from the year 1372; However, my data begins in the year 1471. I think there is an error for the last window of magnitudes. By the way, I had to make some modifications to the plot code because ax._get_lines.prop_cycler is not a valid property in newer versions of Matplotlib. I will attach my modified file.

Could someone explain to me how to correct the last trace corresponding to the magnitude window of 8.8 forward? Thank you.

[Marco Pagani]

Thanks for signaling these problems. My suggestion is to avoid trusting completeness provided by Stepp (or other statistical methods) when you use a limited dataset.

Il giorno gio 13 giu 2024 alle 01:09 Edwar Q S <edwa...@gmail.com> ha scritto:

I already understood what the problem was. Is there a way for my last window to be from 8 to 9 directly?

El Wednesday, June 12, 2024 a la(s) 4:03:24 AM UTC-5, Edwar Q S escribió:

When I apply the stepp 1972 completeness analysis algorithm, I get an anomalous graph for the last magnitude window. As shown in the image, apparently there is data from the year 1372; However, my data begins in the year 1471. I think there is an error for the last window of magnitudes. By the way, I had to make some modifications to the plot code because ax._get_lines.prop_cycler is not a valid property in newer versions of Matplotlib. I will attach my modified file.

Could someone explain to me how to correct the last trace corresponding to the magnitude window of 8.8 forward? Thank you.

--
You received this message because you are subscribed to the Google Groups "OpenQuake Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openquake-use...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openquake-users/b0c2f98f-5cbd-4e7e-8c2d-eac5d74d1f3cn%40googlegroups.com.