WLS for feols() (#273)

* add support for WLS * speed up testing * small documentation updates
py-econometrics · Jan 31, 2024 · 8576859 · 8576859
1 parent 24eafdd
commit 8576859
Show file tree

Hide file tree

Showing 52 changed files with 702 additions and 10,412 deletions.
diff --git a/docs/difference-in-differences.html b/docs/difference-in-differences.html
@@ -147,7 +147,7 @@
     <a class="nav-link" href="./replicating-the-effect.html">Replicating ‘The Effect’ with PyFixest</a>
   </li>  
   <li class="nav-item">
-    <a class="nav-link" href="./reference/index.qmd">Documentation</a>
+    <a class="nav-link" href="./reference/index.html">Documentation</a>
   </li>  
   <li class="nav-item">
     <a class="nav-link" href="./news.html">Changelog</a>

diff --git a/docs/news.html b/docs/news.html
@@ -195,7 +195,7 @@ <h2 class="anchored" data-anchor-id="pyfixest-0.14.0">PyFixest 0.14.0</h2>
 </div>
 <div class="cell-output cell-output-display">
 
-            <div id="LxCHXR"></div>
+            <div id="9KYwVr"></div>
             <script type="text/javascript" data-lets-plot-script="library">
                 if(!window.letsPlotCallQueue) {
                     window.letsPlotCallQueue = [];
@@ -206,7 +206,7 @@ <h2 class="anchored" data-anchor-id="pyfixest-0.14.0">PyFixest 0.14.0</h2>
                 (function() {
                     var script = document.createElement("script");
                     script.type = "text/javascript";
-                    script.src = "https://cdn.jsdelivr.net/gh/JetBrains/lets-plot@v4.2.0/js-package/distr/lets-plot.min.js";
+                    script.src = "https://cdn.jsdelivr.net/gh/JetBrains/lets-plot@v4.0.1/js-package/distr/lets-plot.min.js";
                     script.onload = function() {
                         window.letsPlotCall = function(f) {f();};
                         window.letsPlotCallQueue.forEach(function(f) {f();});
@@ -219,9 +219,9 @@ <h2 class="anchored" data-anchor-id="pyfixest-0.14.0">PyFixest 0.14.0</h2>
                         var div = document.createElement("div");
                         div.style.color = 'darkred';
                         div.textContent = 'Error loading Lets-Plot JS';
-                        document.getElementById("LxCHXR").appendChild(div);
+                        document.getElementById("9KYwVr").appendChild(div);
                     };
-                    var e = document.getElementById("LxCHXR");
+                    var e = document.getElementById("9KYwVr");
                     e.appendChild(script);
                 })()
             </script>

diff --git a/docs/news.qmd b/docs/news.qmd
@@ -1,5 +1,12 @@
 # News
 
+## PyFixest 0.15.0
+
+- Adds support for weighted least squares for `feols()`.
+- Reduces testing time drastically by running tests on fewer random data samples. Qualitatively,
+  the set of test remains identical.
+- Some updates for future `pandas` compatibility.
+
 ## PyFixest 0.14.0
 
 - Moves the documentation to [quartodoc](https://github.com/machow/quartodoc).

diff --git a/docs/reference/demean.demean.qmd b/docs/reference/demean.demean.qmd
@@ -0,0 +1,22 @@
+# demean.demean { #pyfixest.demean.demean }
+
+`demean.demean(x, flist, weights, tol=1e-08, maxiter=100000)`
+
+Workhorse for demeaning an input array `x` based on the specified fixed effects and weights
+via the alternating projections algorithm.
+
+## Parameters
+
+| Name      | Type                | Description                                                                                                    | Default    |
+|-----------|---------------------|----------------------------------------------------------------------------------------------------------------|------------|
+| `x`       | numpy.numpy.ndarray | Input array of shape (n_samples, n_features). Needs to be of type float.                                       | _required_ |
+| `flist`   | numpy.numpy.ndarray | Array of shape (n_samples, n_factors) specifying the fixed effects. Needs to already be converted to integers. | _required_ |
+| `weights` | numpy.numpy.ndarray | Array of shape (n_samples,) specifying the weights.                                                            | _required_ |
+| `tol`     | float               | Tolerance criterion for convergence. Defaults to 1e-08.                                                        | `1e-08`    |
+| `maxiter` | int                 | Maximum number of iterations. Defaults to 100_000.                                                             | `100000`   |
+
+## Returns
+
+| Type                                      | Description                                                                                                                                   |
+|-------------------------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------|
+| typing.Tuple\[numpy.numpy.ndarray, bool\] | A tuple containing the demeaned array of shape (n_samples, n_features) and a boolean indicating whether the algorithm converged successfully. |
diff --git a/reference/detect_singletons.qmd → ...e/detect_singletons.detect_singletons.qmd b/reference/detect_singletons.qmd → ...e/detect_singletons.detect_singletons.qmd
@@ -1,16 +1,4 @@
-# detect_singletons { #pyfixest.detect_singletons }
-
-`detect_singletons`
-
-
-
-## Functions
-
-| Name | Description |
-| --- | --- |
-| [detect_singletons](#pyfixest.detect_singletons.detect_singletons) | Detect singleton fixed effects in a dataset. |
-
-### detect_singletons { #pyfixest.detect_singletons.detect_singletons }
+# detect_singletons.detect_singletons { #pyfixest.detect_singletons.detect_singletons }
 
 `detect_singletons.detect_singletons(ids)`
 
@@ -19,19 +7,19 @@ Detect singleton fixed effects in a dataset.
 This function iterates over the columns of a 2D numpy array representing fixed effects to identify singleton fixed effects.
 An observation is considered a singleton if it is the only one in its group (fixed effect identifier).
 
-#### Parameters
+## Parameters
 
 | Name   | Type                | Description                                                                                                                                                           | Default    |
 |--------|---------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------|------------|
 | `ids`  | numpy.numpy.ndarray | A 2D numpy array representing fixed effects, with a shape of (n_samples, n_features). Elements should be non-negative integers representing fixed effect identifiers. | _required_ |
 
-#### Returns
+## Returns
 
 | Type                | Description                                                                                         |
 |---------------------|-----------------------------------------------------------------------------------------------------|
 | numpy.numpy.ndarray | A boolean array of shape (n_samples,), indicating which observations have a singleton fixed effect. |
 
-#### Notes
+## Notes
 
 The algorithm iterates over columns to identify fixed effects. After each column is processed, it updates the record of
 non-singleton rows. This approach accounts for the possibility that removing an observation in one column can lead to the