Add frequency domain seasonal components to UnobservedComponents #3

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open

jyoder6 wants to merge 202 commits into master from ucm-complex-seas

.gitignore

-Original file line number
+Diff line change
@@ Expand Up / @@ -83,3 +83,7 @@ statsmodels/tsa/statespace/_smoothers/_conventional.pyx @@
     statsmodels/tsa/statespace/_smoothers/_univariate.pyx
     statsmodels/tsa/statespace/_smoothers/_alternative.pyx
     statsmodels/tsa/statespace/_smoothers/_classical.pyx
+    #pytest
+    .cache
+    .pytest_cache

.travis.yml

            
                      Original file line number
                      Diff line number
                      Diff line change
                  
    @@ -69,17 +69,15 @@ notifications:
  
    # Setup anaconda

    before_install:

      - wget http://repo.continuum.io/miniconda/Miniconda-latest-Linux-x86_64.sh -O miniconda.sh

      - wget http://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh -O miniconda.sh

      - chmod +x miniconda.sh

      - ./miniconda.sh -b

      - ./miniconda.sh -b -p /home/travis/miniconda

      - export PATH=/home/travis/miniconda/bin:$PATH

      - export PATH=/home/travis/miniconda2/bin:$PATH

      - export MKL_NUM_THREADS=1

      - export NUMEXPR_NUM_THREADS=1

      - export OMP_NUM_THREADS=1

      - conda config --set always_yes yes

      # Temporarily disabled until conda is fixed

      # - conda update --quiet conda

      - conda update --quiet conda

      # Fix for headless TravisCI

      - "export DISPLAY=:99.0"

      - "sh -e /etc/init.d/xvfb start"

CONTRIBUTING.rst

-Original file line number
+Diff line change
@@ Expand Up @@
     - **One branch. One feature.** Branches are cheap and github makes it easy to merge and delete branches with a few clicks. Avoid the temptation to lump in a bunch of unrelated changes when working on a feature, if possible. This helps us keep track of what has changed when preparing a release.
     - Commit messages should be clear and concise. This means a subject line of less than 80 characters, and, if necessary, a blank line followed by a commit message body. We have an `informal commit format standard <http://www.statsmodels.org/devel/dev/maintainer_notes.html#commit-comments>`_ that we try to adhere to. You can see what this looks like in practice by ``git log --oneline -n 10``. If your commit references or closes a specific issue, you can close it by mentioning it in the `commit message <https://help.github.com/articles/closing-issues-via-commit-messages>`_.  (*For maintainers*: These suggestions go for Merge commit comments too. These are partially the record for release notes.)
-    - Code submissions must always include tests. See our `notes on testing <https://www.statsmodels.org/devel/dev/test_notes.html>`_.
+    - Code submissions must always include tests. See our `notes on testing <http://www.statsmodels.org/devel/dev/test_notes.html>`_.
     - Each function, class, method, and attribute needs to be documented using docstrings. We conform to the `numpy docstring standard <https://github.com/numpy/numpy/blob/master/doc/HOWTO_DOCUMENT.rst.txt#docstring-standard>`_.
     - If you are adding new functionality, you need to add it to the documentation by editing (or creating) the appropriate file in ``docs/source``.
     - Make sure your documentation changes parse correctly. Change into the top-level ``docs/`` directory and type::
@@ Expand Down @@

docs/source/mixed_glm.rst

-Original file line number
+Diff line change
@@ -0,0 +1,33 @@
+    .. currentmodule:: statsmodels.genmod.bayes_mixed_glm
+    Generalized Linear Mixed Effects Models
+    =======================================
+    Generalized Linear Mixed Effects (GLIMMIX) models are generalized
+    linear models with random effects in the linear predictors.
+    Statsmodels currently supports estimation of binomial and Poisson
+    GLIMMIX models using two Bayesian methods: the Laplace approximation
+    to the posterior, and a variational Bayes approximation to the
+    posterior.  Both methods provide point estimates (posterior means) and
+    assessments of uncertainty (posterior standard deviation).
+    The current implementation only supports independent random effects.
+    Technical Documentation
+    -----------------------
+    Unlike Statsmodels mixed linear models, the GLIMMIX implementation is
+    not group-based.  Groups are created by interacting all random effects
+    with a categorical variable.  Note that this creates large, sparse
+    random effects design matrices `exog_vc`.  Internally, `exog_vc` is
+    converted to a scipy sparse matrix.  When passing the arguments
+    directly to the class initializer, a sparse matrix may be passed.
+    When using formulas, a dense matrix is created then converted to
+    sparse.  For very large problems, it may not be feasible to use
+    formulas due to the size of this dense intermediate matrix.
+    References
+    ^^^^^^^^^^
+    Blei, Kucukelbir, McAuliffe (2017).  Variational Inference: A review
+    for Statisticians https://arxiv.org/pdf/1601.00670.pdf

docs/source/release/version0.9.rst

-Original file line number
+Diff line change
@@ Expand Up / @@ -28,6 +28,13 @@ are mentioned in the docstrings. @@
     The following major new features appear in this version.
+    Generalized linear mixed models
+    -------------------------------
+    Limited support for GLIMMIX models is now included in the genmod
+    module.  Binomial and Poisson models with independent random effects
+    can be fit using Bayesian methods (Laplace and mean field
+    approximations to the posterior).
     Documentation
     -------------
@@ Expand All / @@ -37,6 +44,8 @@ Documentation @@
     Other important improvements
     ----------------------------
+    * MICE (multiple imputation) can use regularized model fitters in the
+      imputation step.
@@ Expand All @@
     for bugs that cause that silently incorrect numbers are returned.
     https://github.com/statsmodels/statsmodels/issues?q=label%3Atype-bug-wrong+is%3Aclosed
+    * Refitting elastic net regularized models using the `refit=True`
+      option now returns the unregularized parameters for the coefficients
+      selected by the regularized fitter, as documented. #4213
     Backwards incompatible changes and deprecations
@@ Expand Down @@

examples/notebooks/distributed_estimation.ipynb

            
                      Original file line number
                      Diff line number
                      Diff line change
                  
    @@ -9,13 +9,12 @@
  
      },

      {

       "cell_type": "code",

       "execution_count": 1,

       "metadata": {

        "collapsed": false

       },

       "execution_count": null,

       "metadata": {},

       "outputs": [],

       "source": [

        "import numpy as np\n",

        "from scipy.stats.distributions import norm\n",

        "from statsmodels.base.distributed_estimation import DistributedModel\n",

        "\n",

        "def _exog_gen(exog, partitions):\n",

    @@ -52,16 +51,14 @@
  
      },

      {

       "cell_type": "code",

       "execution_count": 2,

       "metadata": {

        "collapsed": true

       },

       "execution_count": null,

       "metadata": {},

       "outputs": [],

       "source": [

        "X = np.random.normal(size=(1000, 25))\n",

        "beta = np.random.normal(size=25)\n",

        "beta *= np.random.randint(0, 2, size=25)\n",

        "y = X.dot(beta) + np.random.normal(size=1000)\n",

        "y = norm.rvs(loc=X.dot(beta))\n",

        "m = 5"

       ]

      },

    @@ -74,10 +71,8 @@
  
      },

      {

       "cell_type": "code",

       "execution_count": 3,

       "metadata": {

        "collapsed": false

       },

       "execution_count": null,

       "metadata": {},

       "outputs": [],

       "source": [

        "debiased_OLS_mod = DistributedModel(m)\n",

    @@ -94,17 +89,17 @@
  
      },

      {

       "cell_type": "code",

       "execution_count": 4,

       "execution_count": null,

       "metadata": {

        "collapsed": false

        "scrolled": false

       },

       "outputs": [],

       "source": [

        "from statsmodels.genmod.generalized_linear_model import GLM\n",

        "from statsmodels.genmod.families import Binomial\n",

        "from statsmodels.genmod.families import Gaussian\n",

        "\n",

        "debiased_GLM_mod = DistributedModel(m, model_class=GLM,\n",

        "                                    init_kwds={\"family\": Binomial()})\n",

        "                                    init_kwds={\"family\": Gaussian()})\n",

        "debiased_GLM_fit = debiased_GLM_mod.fit(zip(_endog_gen(y, m), _exog_gen(X, m)),\n",

        "                                        fit_kwds={\"alpha\": 0.2})"

       ]

    @@ -118,10 +113,8 @@
  
      },

      {

       "cell_type": "code",

       "execution_count": 5,

       "metadata": {

        "collapsed": true

       },

       "execution_count": null,

       "metadata": {},

       "outputs": [],

       "source": [

        "from statsmodels.base.distributed_estimation import _est_regularized_naive, _join_naive\n",

    @@ -142,10 +135,8 @@
  
      },

      {

       "cell_type": "code",

       "execution_count": 6,

       "metadata": {

        "collapsed": true

       },

       "execution_count": null,

       "metadata": {},

       "outputs": [],

       "source": [

        "from statsmodels.base.distributed_estimation import _est_unregularized_naive, DistributedResults\n",

    @@ -175,7 +166,7 @@
  
       "name": "python",

       "nbconvert_exporter": "python",

       "pygments_lexer": "ipython2",

       "version": "2.7.10"

       "version": "2.7.14"

      }

     },

     "nbformat": 4,

examples/notebooks/regression_diagnostics.ipynb

-Original file line number
+Diff line change
@@ Expand Up / @@ -46,7 +46,7 @@ @@
         "url = 'http://vincentarelbundock.github.io/Rdatasets/csv/HistData/Guerry.csv'\n",
         "dat = pd.read_csv(url)\n",
         "\n",
-        "# Fit regression model (using the natural log of one of the regressaors)\n",
+        "# Fit regression model (using the natural log of one of the regressors)\n",
         "results = smf.ols('Lottery ~ Literacy + np.log(Pop1831)', data=dat).fit()\n",
         "\n",
         "# Inspect the results\n",
@@ Expand Down Expand Up / @@ -258,4 +258,4 @@ @@
      },
      "nbformat": 4,
      "nbformat_minor": 0
-    }
+    }

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add frequency domain seasonal components to UnobservedComponents #3

Uh oh!

Diff view

Diff view

There are no files selected for viewing

Uh oh!

Add frequency domain seasonal components to UnobservedComponents #3

Are you sure you want to change the base?

Uh oh!

Add frequency domain seasonal components to UnobservedComponents #3

Uh oh!

Uh oh!

Diff view

Diff view

There are no files selected for viewing

Uh oh!