MAINT: Update anaconda, README, and bibtex style (#23)

* update ga workflow version

* adjust formatting in bayes_nonconj

* upgrade to anaconda=2024.02

* default README format

* enable author-year bibtex style

* tmp: disable build cache

.github/workflows/cache.yml

@@ -27,7 +27,7 @@ jobs:
     needs: deploy-runner
     runs-on: [self-hosted, cml-gpu]
     container:
-      image: docker://mmcky/quantecon-lecture-python:cuda-12.3.1-anaconda-2023-09-py311
+      image: docker://mmcky/quantecon-lecture-python:cuda-12.3.1-anaconda-2024-02-py311
       options: --gpus all
     steps:
       - uses: actions/checkout@v4

.github/workflows/ci.yml

@@ -24,7 +24,7 @@ jobs:
     needs: deploy-runner
     runs-on: [self-hosted, cml-gpu]
     container:
-      image: docker://mmcky/quantecon-lecture-python:cuda-12.3.1-anaconda-2023-09-py311
+      image: docker://mmcky/quantecon-lecture-python:cuda-12.3.1-anaconda-2024-02-py311
       options: --gpus all
     steps:
       - uses: actions/checkout@v4
@@ -40,13 +40,13 @@ jobs:
       - name: Display Pip Versions
         shell: bash -l {0}
         run: pip list
-      - name: Download "build" folder (cache)
-        uses: dawidd6/action-download-artifact@v2
-        with:
-          workflow: cache.yml
-          branch: main
-          name: build-cache
-          path: _build
+      # - name: Download "build" folder (cache)
+      #   uses: dawidd6/action-download-artifact@v2
+      #   with:
+      #     workflow: cache.yml
+      #     branch: main
+      #     name: build-cache
+      #     path: _build
       # Build Assets (Download Notebooks and PDF via LaTeX)
       - name: Build Download Notebooks (sphinx-tojupyter)
         shell: bash -l {0}

.github/workflows/linkcheck.yml

@@ -18,7 +18,7 @@ jobs:
       - name: Checkout
         uses: actions/checkout@v4
       - name: Setup Anaconda
-        uses: conda-incubator/setup-miniconda@v2
+        uses: conda-incubator/setup-miniconda@v3
         with:
           auto-update-conda: true
           auto-activate-base: true

.github/workflows/publish.yml

@@ -28,7 +28,7 @@ jobs:
     needs: deploy-runner
     runs-on: [self-hosted, cml-gpu]
     container:
-      image: docker://mmcky/quantecon-lecture-python:cuda-12.3.1-anaconda-2023-09-py311
+      image: docker://mmcky/quantecon-lecture-python:cuda-12.3.1-anaconda-2024-02-py311
       options: --gpus all
     steps:
       - name: Checkout

README.md

@@ -1,3 +1,16 @@
-# lecture-stats
+# Statistics for Computational Economics
 
-Statistics for Computational Economics
+This website presents a set of lectures on statistics for computational economics.
+
+## Jupyter notebooks
+
+Jupyter notebook versions of each lecture are available for download
+via the website.
+
+## Contributions
+
+To comment on the lectures please add to or open an issue in the issue tracker (see above).
+
+We welcome pull requests!  
+
+Please read the [QuantEcon style guide](https://manual.quantecon.org/intro.html) first, so that you can match our style.

environment.yml

@@ -3,7 +3,7 @@ channels:
   - default
 dependencies:
   - python=3.11
-  - anaconda=2023.09
+  - anaconda=2024.02
   - pip
   - pip:
     - jupyter-book==0.15.1

lectures/_config.yml

@@ -34,6 +34,7 @@ latex:
 sphinx:
   extra_extensions: [sphinx_multitoc_numbering, sphinxext.rediraffe, sphinx_tojupyter, sphinxcontrib.youtube, sphinx.ext.todo, sphinx_exercise, sphinx_togglebutton, sphinx.ext.intersphinx, sphinx_reredirects]
   config:
+    bibtex_reference_style: author_year
     nb_mime_priority_overrides: [
        # HTML
        ['html', 'application/vnd.jupyter.widget-view+json', 10],

lectures/bayes_nonconj.md

@@ -24,7 +24,7 @@ Having a likelihood and prior that  are conjugate can simplify calculation of a
 
 But in many situations  the likelihood and prior need not form a conjugate pair.
 
- - after all, a person's prior is his or her own business and would take a form conjugate to a likelihood only by remote coincidence
+- after all, a person's prior is his or her own business and would take a form conjugate to a likelihood only by remote coincidence
 
 In these situations, computing a posterior can become very challenging.
 
@@ -78,7 +78,7 @@ We use `numpyro` with assistance from  `jax` to approximate a  posterior distrib
 
 We use several alternative prior distributions.
 
-We  compare computed posteriors  with ones associated with a conjugate prior as described in  {doc}`the quantecon lecture <prob_meaning>`.
+We compare computed posteriors  with ones associated with a conjugate prior as described in  {doc}`the quantecon lecture <prob_meaning>`.
 
 
 ### Analytical Posterior
@@ -119,8 +119,6 @@ $$
 =\frac{(1 -\theta)^{\beta+N-k-1} \theta^{\alpha+k-1}}{\int_0^1 (1 - \theta)^{\beta+N-k-1} \theta^{\alpha+k-1} d\theta} .
 $$
 
-
-
 Thus,
 
 $$
@@ -191,7 +189,7 @@ By paying the cost of restricting the putative posterior to have a restricted fu
 the problem of approximating a posteriors is transformed to a well-posed optimization problem that seeks parameters of the putative posterior  that minimize
 a Kullback-Leibler (KL) divergence between true posterior and the putatitive posterior  distribution.
 
-  - minimizing the KL divergence is  equivalent with  maximizing a criterion called  the **Evidence Lower Bound** (ELBO), as we shall verify soon.
+- minimizing the KL divergence is  equivalent with  maximizing a criterion called  the **Evidence Lower Bound** (ELBO), as we shall verify soon.
 
 ## Prior Distributions
 
@@ -277,14 +275,14 @@ $$
 p\left(Y\right)=\int d\theta p\left(Y\mid\theta\right)p\left(Y\right).
 $$ (eq:intchallenge)
 
-The integral on the right side of {eq}`eq:intchallenge`  is typically difficult to compute.
+The integral on the right side of {eq}`eq:intchallenge` is typically difficult to compute.
 
 Consider a  **guide distribution** $q_{\phi}(\theta)$ parameterized by $\phi$ that we'll use to approximate the posterior.
 
 We choose  parameters $\phi$ of the guide distribution to minimize a Kullback-Leibler (KL)  divergence between the approximate posterior $q_{\phi}(\theta)$ and  the posterior:
 
 $$
- D_{KL}(q(\theta;\phi)\;\|\;p(\theta\mid Y)) \equiv -\int d\theta q(\theta;\phi)\log\frac{p(\theta\mid Y)}{q(\theta;\phi)}
+D_{KL}(q(\theta;\phi)\;\|\;p(\theta\mid Y)) \equiv -\int d\theta q(\theta;\phi)\log\frac{p(\theta\mid Y)}{q(\theta;\phi)}
 $$
 
 Thus, we want a **variational distribution** $q$ that solves
@@ -296,7 +294,8 @@ $$
 Note that
 
 $$
-\begin{aligned}D_{KL}(q(\theta;\phi)\;\|\;p(\theta\mid Y)) & =-\int d\theta q(\theta;\phi)\log\frac{P(\theta\mid Y)}{q(\theta;\phi)}\\
+\begin{aligned}
+D_{KL}(q(\theta;\phi)\;\|\;p(\theta\mid Y)) & =-\int d\theta q(\theta;\phi)\log\frac{P(\theta\mid Y)}{q(\theta;\phi)}\\
  & =-\int d\theta q(\theta)\log\frac{\frac{p(\theta,Y)}{p(Y)}}{q(\theta)}\\
  & =-\int d\theta q(\theta)\log\frac{p(\theta,Y)}{p(\theta)q(Y)}\\
  & =-\int d\theta q(\theta)\left[\log\frac{p(\theta,Y)}{q(\theta)}-\log p(Y)\right]\\