[docs] Clarify that describe_distribution() returns excess kurtosis

[Copilot]

Summary

This PR clarifies in the documentation that describe_distribution() returns excess kurtosis rather than raw kurtosis values.

Background

As raised in the issue, there was potential confusion about the kurtosis values returned by describe_distribution(). Since a normal distribution has a kurtosis of 3, but describe_distribution() shows a kurtosis value near 0 for normal distributions, it's important to clarify that the function returns excess kurtosis (kurtosis - 3).

The kurtosis() function already mentions "(Excess)" in its title and documentation, but describe_distribution() didn't make this clear.

Changes

Updated the documentation in R/describe_distribution.R:

1. Modified the function description to explicitly mention "(excess) kurtosis":

   # Before:
   # centrality, dispersion, range, skewness, kurtosis
   
   # After:
   # centrality, dispersion, range, skewness, (excess) kurtosis

2. Added a @seealso section linking to the kurtosis() function for users who want more details about the implementation and interpretation.

Impact

This documentation improvement helps users correctly interpret the kurtosis values:

• Excess kurtosis ≈ 0: Mesokurtic distribution (like normal distribution)
• Excess kurtosis > 0: Leptokurtic distribution (fatter tails)
• Excess kurtosis < 0: Platykurtic distribution (thinner tails)

Closes the issue by providing the clarity requested.

Original prompt

---

This section details on the original issue you should resolve

<issue_title>[docs] clarify kurtosis column: is it excess kurtosis?</issue_title>
<issue_description>

I heard that the kurtosis of a normal distribution is 3, and excess kurtosis is calculated as kurtosis - 3. If that's the case, worth clarifying in the docs that the kurtosis col corresponds to excess kurtosis</issue_description>

<agent_instructions>Edit the description of describe_distribution() to mention that it is "(excess)" kurtosis</agent_instructions>

Comments on the Issue (you are @copilot in this section)

@strengejacke Any references we could add despite mysterious voices in your head? @mattansb It's already in the title:

datawizard/R/skewness_kurtosis.R

Line 1 in 3e29512

#' Compute Skewness and (Excess) Kurtosis

You can use the same reference of:

D. N. Joanes and C. A. Gill (1998). Comparing measures of sample skewness and kurtosis. The Statistician, 47, 183–189.</comment_new>
<comment_new>@strengejacke
And the docs also explain how to interpret that values. So nothing to do here?</comment_new>
<comment_new>@DominiqueMakowski
My bad! cheers</comment_new>
<comment_new>@DominiqueMakowski
ah my bad x2 I have misread the original issue, the issue was that "the kurtosis function itself describes how to interpet it, but the describe_distribution() doesn't" 😁

I think it's just a matter of amending the description:

This function describes a distribution by a set of indices (e.g., measures of centrality, dispersion, range, skewness, kurtosis).

To mention "(excess)" kurtosis here and a link to the corresponding function

Fixes #661

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.

[DominiqueMakowski]

lgtm