> For the complete documentation index, see [llms.txt](https://docs.everesteer.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.everesteer.ai/scoring/corr.md).

# CORR

CORR is not a Spearman coefficient. Per exped it is Pearson correlation after a specific transform, then averaged across the round.

{% stepper %}
{% step %}

### Align

Predictions and the realised target are aligned on the ids that exist in both. Too few overlapping ids, or a side with no variance, scores as empty on that exped. The kernel does not invent a number.
{% endstep %}

{% step %}

### Transform predictions

Rank-gaussianize **predictions only** (average ranks, then the inverse normal), then apply a signed power transform.
{% endstep %}

{% step %}

### Transform the target

Center the **target**. Do not rank-gaussianize it. Apply the same signed power.
{% endstep %}

{% step %}

### Pearson

Take Pearson of those two series. That per-exped value is the CORR term of the round score.
{% endstep %}
{% endstepper %}

{% columns %}
{% column %}
{% hint style="info" %}
**Partial.** N-day forward returns, rank-normalised across the cross-section.
{% endhint %}
{% endcolumn %}

{% column %}
{% hint style="success" %}
**Resolved.** The mature quintile-binned `target_everest_20`.
{% endhint %}
{% endcolumn %}
{% endcolumns %}

Those are different series. A partial CORR is not CORR against the mature target.

CORR is also the fallback order for entries that cannot yet be scored, and the whole-board fallback when nothing on a board can be scored. It is not, by itself, what a scored board ranks on.

The `everestapi[scoring]` extra ships a faithful re-implementation you can run on a hold-out. Official round scores are still server-side, against an answer key you never receive. See Offline scoring toolkit.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.everesteer.ai/scoring/corr.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.