Propose to add _"miscalibration"_ plot

[mafed]

Instead of plotting y vs. x in the calibration plot, drawing y–x vs. x gives a miscalibration plot, which is focusing on what is most interesting: deviation from equality. I think it's an easy modification and might be worth it.

[proshano]

I like this idea!

---

From: Federico Mattiello @.> Sent: Sunday, August 27, 2023 9:56:46 AM To: BavoDC/CalibrationCurves @.> Cc: Subscribed @.***> Subject: [BavoDC/CalibrationCurves] Propose to add "miscalibration" plot (Issue #6)

Instead of plotting y vs. x in the calibration plot, drawing y–x vs. x gives a miscalibration plot, which is focusing on what is most interesting: deviation from equality. I think it's an easy modification and might be worth it.

— Reply to this email directly, view it on GitHubhttps://github.com/BavoDC/CalibrationCurves/issues/6, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AEJTO3GC6ZRMOFFCSIUVSPTXXNGZ5ANCNFSM6AAAAAA4AKVFU4. You are receiving this because you are subscribed to this thread.Message ID: @.***>

[BavoDC]

Hi Federico,

This sounds like an interesting idea! I haven't heard of it before, but it definitely sounds useful. I added it to my to-do list.

Which references would you suggest for the miscalibration plot? I can then look at it in detail.

Best,

Bavo.

[mafed]

Hi Bavo, I actually don't have any reference for this. It's just an idea I had during a project where I was validating a model. It shifts the focus to what we're really looking at: deviations from the bisector line.

Cheers, Federico

On Wed, 30 Aug 2023, 14:58 Bavo (De Cock) Campo, @.***> wrote:

Hi Federico,

This sounds like an interesting idea! I haven't heard of it before, but it definitely sounds useful. I added it to my to-do list.

Which references would you suggest for the miscalibration plot? I can then look at it in detail.

Best,

Bavo.

— Reply to this email directly, view it on GitHub https://github.com/BavoDC/CalibrationCurves/issues/6#issuecomment-1699121637, or unsubscribe https://github.com/notifications/unsubscribe-auth/AATE7LI6SSYP7FBEJVVZA3DXX42IPANCNFSM6AAAAAA4AKVFU4 . You are receiving this because you authored the thread.Message ID: @.***>

[BavoDC]

Hi Federico,

Thanks again for the suggestion. Unfortunately, due to lack of time, I will have to postpone it until later.

Best,

Bavo.