Missingness support for Simulators

docs/tutorials/gentle_intro/00_index.ipynb

@@ -1,60 +1,64 @@
 {
-  "cells": [
-    {
-      "cell_type": "markdown",
-      "metadata": {},
-      "source": [
-        "# A Gentle Introduction to Dynestyx\n",
-        "\n",
-        "This multi-part tutorial is a conceptual complement to the [Quickstart](../../quickstart.ipynb). We start from basic Bayesian inference in NumPyro, then show how **dynestyx** extends it to sample and infer **dynamical systems**—with a clear separation between *what* the model is (parameters + dynamics + observations) and *how* we simulate or compute likelihoods (simulators and filters)."
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {},
-      "source": [
-        "## Contents\n",
-        "\n",
-        "1. **[Part 1: NumPyro and the Bayesian workflow](../01_numpyro_bayesian_workflow/)** — Linear regression with NUTS and ArviZ; generating data with `Predictive`; prior and posterior predictive (Bayesian workflow).\n",
-        "\n",
-        "2. **[Part 2: Dynestyx — discrete-time dynamical systems](../02_dynestyx_discrete_intro/)** — What dynestyx is; separation of concerns; a first model (discrete-time stochastic volatility); context and handlers; generating data with `DiscreteTimeSimulator`; NUTS + simulator for $p(x, \\theta \\mid \\text{data})$.\n",
-        "\n",
-        "3. **[Part 3: Filtering and the marginal log-likelihood](../03_filtering_mll/)** — Computing MLL with the particle filter at $\\theta_{\\text{mean}}$ and $\\theta_{\\text{true}}$; switching to Taylor KF; canonical reference placeholder.\n",
-        "\n",
-        "4. **[Part 4: Filtering + NUTS — pseudomarginal inference](../04_filtering_nuts_pseudomarginal/)** — Using the filter’s MLL inside NUTS to sample only parameters; when this is “overkill” vs. valuable (with links to be filled).\n",
-        "\n",
-        "5. **[Part 5: SVI and warming up NUTS](../05_svi/)** — SVI as a fast alternative; using SVI to initialize NUTS; execution times; rapid prototyping.\n",
-        "\n",
-        "6. **[Part 6a: Continuous-time dynamical systems (SDEs)](../06_continuous_time/)** — `ContinuousTimeStateEvolution` (drift, diffusion coefficient, diffusion covariance); `SDESimulator`; Lorenz 63 with partial observations; `LinearGaussianObservation` and $H$; NUTS + EnKF; full-observations deep dive link.\n",
-        "\n",
-        "7. **[Part 6b: Continuous-time dynamical systems (ODEs)](../06b_odes/)** - `ContinuousTimeStateEvolution` (drift and no diffusions); `ODESimulator`. Probabilistic numerics coming soon!\n",
-        "\n",
-        "8. **[Part 7: Hidden Markov Models (HMMs)](../07_hmm/)** — Working with categorical state spaces models (filtering and Bayesian parameter estimation).\n",
-        "\n",
-        "9. **[Part 8: Hierarchical / mixed-effect inference](../08_hierarchical_inference/)** — `plate` for trajectory-level parameters; hierarchical dynamical models and inference.\n",
-        "\n",
-        "10. **[Part 9: Discrete-time smoothing](../09_discrete_smoothing/)** — `Smoother` for discrete-time models; smoothed state summaries and future-only posterior rollout.\n",
-        "\n",
-        "11. **[Part 10: Continuous-time smoothing](../10_continuous_smoothing/)** — Continuous-discrete smoothing; smoothed state summaries and future-only posterior rollout."
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {},
-      "source": []
-    }
-  ],
-  "metadata": {
-    "kernelspec": {
-      "display_name": "Python 3",
-      "language": "python",
-      "name": "python3"
-    },
-    "language_info": {
-      "name": "python",
-      "version": "3.12.0"
-    }
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# A Gentle Introduction to Dynestyx\n",
+    "\n",
+    "This multi-part tutorial is a conceptual complement to the [Quickstart](../../quickstart.ipynb). We start from basic Bayesian inference in NumPyro, then show how **dynestyx** extends it to sample and infer **dynamical systems**—with a clear separation between *what* the model is (parameters + dynamics + observations) and *how* we simulate or compute likelihoods (simulators and filters)."
+   ]
   },
-  "nbformat": 4,
-  "nbformat_minor": 4
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Contents\n",
+    "\n",
+    "1. **[Part 1: NumPyro and the Bayesian workflow](../01_numpyro_bayesian_workflow/)** — Linear regression with NUTS and ArviZ; generating data with `Predictive`; prior and posterior predictive (Bayesian workflow).\n",
+    "\n",
+    "2. **[Part 2: Dynestyx — discrete-time dynamical systems](../02_dynestyx_discrete_intro/)** — What dynestyx is; separation of concerns; a first model (discrete-time stochastic volatility); context and handlers; generating data with `DiscreteTimeSimulator`; NUTS + simulator for $p(x, \\theta \\mid \\text{data})$.\n",
+    "\n",
+    "3. **[Part 3: Filtering and the marginal log-likelihood](../03_filtering_mll/)** — Computing MLL with the particle filter at $\\theta_{\\text{mean}}$ and $\\theta_{\\text{true}}$; switching to Taylor KF; canonical reference placeholder.\n",
+    "\n",
+    "4. **[Part 4: Filtering + NUTS — pseudomarginal inference](../04_filtering_nuts_pseudomarginal/)** — Using the filter’s MLL inside NUTS to sample only parameters; when this is “overkill” vs. valuable (with links to be filled).\n",
+    "\n",
+    "5. **[Part 5: SVI and warming up NUTS](../05_svi/)** — SVI as a fast alternative; using SVI to initialize NUTS; execution times; rapid prototyping.\n",
+    "\n",
+    "6. **[Part 6a: Continuous-time dynamical systems (SDEs)](../06_continuous_time/)** — `ContinuousTimeStateEvolution` (drift, diffusion coefficient, diffusion covariance); `SDESimulator`; Lorenz 63 with partial observations; `LinearGaussianObservation` and $H$; NUTS + EnKF; full-observations deep dive link.\n",
+    "\n",
+    "7. **[Part 6b: Continuous-time dynamical systems (ODEs)](../06b_odes/)** - `ContinuousTimeStateEvolution` (drift and no diffusions); `ODESimulator`. Probabilistic numerics coming soon!\n",
+    "\n",
+    "8. **[Part 7: Hidden Markov Models (HMMs)](../07_hmm/)** — Working with categorical state spaces models (filtering and Bayesian parameter estimation).\n",
+    "\n",
+    "9. **[Part 8: Hierarchical / mixed-effect inference](../08_hierarchical_inference/)** — `plate` for trajectory-level parameters; hierarchical dynamical models and inference.\n",
+    "\n",
+    "10. **[Part 9: Discrete-time smoothing](../09_discrete_smoothing/)** — `Smoother` for discrete-time models; smoothed state summaries and future-only posterior rollout.\n",
+    "\n",
+    "11. **[Part 10: Continuous-time smoothing](../10_continuous_smoothing/)** — Continuous-discrete smoothing; smoothed state summaries and future-only posterior rollout.\n",
+    "\n",
+    "12. **[Part 11: Missing observations with Filters and Smoothers](../11_missing_observations/)** — Whole-row and partial missingness with the cuthbert Kalman filter/smoother path.\n",
+    "\n",
+    "13. **[Part 11b: Missing observations with `Simulator` + MCMC](../11b_missing_observations_simulator_mcmc/)** — Joint posterior inference over parameters and latent states under Gaussian missing-data simulator conditioning."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "name": "python",
+   "version": "3.12.0"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
 }