[qr_decomp.md] Update np.random → Generator API

[Chihiro2000GitHub]

Summary

This PR migrates legacy NumPy random API usage in qr_decomp.md as part of QuantEcon/meta#299.

The four module-level np.random.* calls are replaced with calls on an explicit generator rng = np.random.default_rng(), defined once in the lecture's import cell and reused throughout.

Related PRs and issues

I checked for open PRs and issues related to this lecture. No open PR modifies qr_decomp.md. The one related issue is #582, on seeding lectures with stochastic output — see the note for reviewers below.

Details

• rng = np.random.default_rng() added to the import cell (import numpy as np / from scipy.linalg import qr).
• np.random.random(...) → rng.random(...) in the QR eigenvalue experiment and in the PCA section's random moments.
• np.random.multivariate_normal(...) → rng.multivariate_normal(...).
• No fixed seed was introduced, since the lecture did not seed before.
• No prose changes were needed: the text does not refer to np.random as a module-level function, and the displayed results are eigenvalue comparisons and np.abs(...).max() residuals rather than specific realized draws.
• The lecture contains no Numba (@jit, @njit, @jitclass, parallel=True, prange) code.
• A full local build completed successfully and qr_decomp.md executed without errors.

Note for reviewers

This lecture produces stochastic output and has never set a seed, which is the situation tracked in #582. I have kept it unseeded here to stay within the scope of the API migration, but please feel free to add a seed as part of that issue if you would prefer.

Hi @mmcky and @HumphreyYang, I'd be grateful if you could take a look when you have time.

[github-actions]

📖 Netlify Preview Ready!

Preview URL: https://pr-999--sunny-cactus-210e3e.netlify.app

Commit: 541c3bf

📚 Changed Lectures

• qr_decomp

---

Build Info

• Workflow: Build Project [using jupyter-book]