Multi yoy integration - combines PRs #456, #460 and #464 at once.

[cdeline]

See individual PR's #456, #460 and #464 for discussion.

Issues & PRs

• Closes 394 multi yoy - allows multiple slope calculations across multi-year periods #464
• Closes 459 yoy label='center' #460
• Closes Degradation timeseries plot - add left or right label option #456
• Closes degradation_year_on_year should allow center-labeled YoY_values #459
• Closes degradation_timeseries_plot should have different label= options #455
• Closes Multi-YoY analysis based on Hugo Quest et al 2023 #394

• Code changes are covered by tests
• Code changes have been evaluated for compatibility/integration with TrendAnalysis
• New functions added to __init__.py
• API.rst is up to date, along with other sphinx docs pages
• Example notebooks are rerun and differences in results scrutinized
• Updated changelog

[martin-springer]

users would need to pass label=None explicitly to trigger those checks but end up with label='right'. I think it's best to rely on the default handling.

I'll remove the label is None handling in degradation.py and plotting.py. I'll also update the tests that specifically trigger this. Commit 0fe8ebd

[Copilot]

Pull request overview

Copilot reviewed 10 out of 11 changed files in this pull request and generated 5 comments.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[martin-springer]

With pandas three now released, should we think about dropping support for pandas 1.x.x?

[cdeline]

yeah, maybe we start with a deprecation warning for pandas < 2.0 in our next release. Pandas 2.0 was only put out in 2023 so we can probably hang on to this functionality for a few more years - its not really causing any harm. PVLib is still on pandas > 1.3.3 so we should probably still support it too...

[cdeline]

@martin-springer - check out the example notebook with label='center' and multi_yoy=True functionality.

[martin-springer]

@martin-springer - check out the example notebook with label='center' and multi_yoy=True functionality.

@cdeline - It's looking great. The multi-yoy analysis really narrows down the distribution!

Regarding the labelling:

1. Should we add an index name to the tables:

?? some name here ??	dt_right	dt_center	dt_left
2010-08-27 12:00:00-07:00	2011-02-26 00:00:00-07:00	2010-08-27 12:00:00-07:00	2010-02-26 00:00:00-07:00

2. Should we try to come up with a plot that illustrates the difference?

[cdeline]

Regarding the labelling:

1. Should we add an index name to the tables:

?? some name here ?? dt_right dt_center dt_left
2010-08-27 12:00:00-07:00 2011-02-26 00:00:00-07:00 2010-08-27 12:00:00-07:00 2010-02-26 00:00:00-07:00

previous label was 'dt'. We could go back to that, or have a more descriptive label of either dt_right, dt_center or dt_left. I'm not sure how much of a problem it would be to have a returned dataframe with index name that changes depending on what labeling is selected. Keeping 'dt' index label would be consistent with previous version.

2. Should we try to come up with a plot that illustrates the difference?

These are two illustrations that I've used in the past to describe center vs right labeling. I could just upload the images to the notebook as an example:

[martin-springer]

@cdeline - great, thank you for making those changes!