DOC: promote tutorial section headers so they show up in the index

docs/notebooks/tutorial/blp.ipynb

@@ -37,19 +37,14 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "In this tutorial, we'll use data from :ref:`references:Berry, Levinsohn, and Pakes (1995)` to solve the paper's automobile problem.\n",
-    "\n",
-    "\n",
-    "## Adding a Supply Side with Automobile Data\n",
-    "\n",
-    "This tutorial is similar to the [fake cereal tutorial](nevo.ipynb), but exhibits some other features of pyblp:\n",
+    "In this tutorial, we'll use data from :ref:`references:Berry, Levinsohn, and Pakes (1995)` to solve the paper's automobile problem. This tutorial is similar to the [fake cereal tutorial](nevo.ipynb), but exhibits some other features of pyblp:\n",
     "\n",
     "- Incorporating a supply side into demand estimation.\n",
     "- Allowing for simple price-income demographic effects.\n",
     "- Calculating clustered standard errors.\n",
     "\n",
     "\n",
-    "### Loading the Data\n",
+    "## Loading Data\n",
     "\n",
     "We'll use [pandas](https://pandas.pydata.org/) to load two sets of data:\n",
     "\n",
@@ -418,7 +413,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### Setting up the Problem\n",
+    "## Setting up the Problem\n",
     "\n",
     "Unlike the fake cereal problem, we won't absorb any fixed effects in the automobile problem, so the linear part of demand $X_1$ has more components. We also need to specify a formula for the random coefficients $X_2$, including a random coefficient on the constant, which captures correlation among all inside goods.\n",
     "\n",
@@ -560,7 +555,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### Solving the Problem"
+    "## Solving the Problem"
    ]
   },
   {
@@ -707,7 +702,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.15"
+   "version": "3.9.15"
   },
   "pycharm": {
    "stem_cell": {

docs/notebooks/tutorial/nevo.ipynb

@@ -69,7 +69,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Application of Random Coefficients Logit with the Fake Cereal Data\n",
+    "## Random Coefficients\n",
     "\n",
     "To include random coefficients we need to add a :class:`Formulation` for the demand-side nonlinear characteristics $X_2$.\n",
     "\n",
@@ -112,7 +112,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### Specification of Random Taste Parameters\n",
+    "## Specification of Random Taste Parameters\n",
     "\n",
     "It is common to assume that $f(\\beta_i \\mid \\theta)$ follows a multivariate normal distribution, and to break it up into three parts:\n",
     "\n",
@@ -133,7 +133,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### Loading the Data\n",
+    "## Loading Data\n",
     "\n",
     "The `product_data` argument of :class:`Problem` should be a structured array-like object with fields that store data. Product data can be a structured [NumPy](https://numpy.org/) array, a [pandas](https://pandas.pydata.org/) DataFrame, or other similar objects."
    ]
@@ -375,7 +375,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### Setting Up and Solving the Problem Without Demographics\n",
+    "## Setting Up and Solving the Problem Without Demographics\n",
     "\n",
     "Formulations, product data, and an integration configuration are collectively used to initialize a :class:`Problem`. Once initialized, :meth:`Problem.solve` runs the estimation routine. The arguments to :meth:`Problem.solve` configure how estimation is performed. For example, `optimization` and `iteration` arguments configure the optimization and iteration routines that are used by the outer and inner loops of estimation.\n",
     "\n",
@@ -761,7 +761,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### Adding Demographics to the Problem\n",
+    "## Adding Demographics to the Problem\n",
     "\n",
     "To add demographic data we need to make two changes:\n",
     "\n",
@@ -1094,7 +1094,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### Restricting Parameters\n",
+    "## Restricting Parameters\n",
     "\n",
     "Because the interactions between `price`, `income`, and `income_squared` are potentially collinear, we might worry that $\\hat{\\Pi}_{21} = 588$ and  $\\hat{\\Pi}_{22} = -30.2$ are pushing the price coefficient in opposite directions. Both are large in magnitude but statistically insignficant. One way of dealing with this is to restrict $\\Pi_{22} = 0$.\n",
     "\n",
@@ -1275,7 +1275,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.8"
+   "version": "3.9.15"
   },
   "pycharm": {
    "stem_cell": {

docs/notebooks/tutorial/petrin.ipynb

@@ -55,15 +55,10 @@
    "id": "7556f474",
    "metadata": {},
    "source": [
-    "In this tutorial, we'll use data from :ref:`references:Petrin (2002)` to solve the paper's automobile problem.\n",
+    "In this tutorial, we'll use data from :ref:`references:Petrin (2002)` to solve the paper's automobile problem. This tutorial is similar to the [first automobile tutorial](blp.ipynb), but exhibits how to incorporate micro moments into estimation.\n",
     "\n",
     "\n",
-    "## Adding Micro Moments with Automobile Data\n",
-    "\n",
-    "This tutorial is similar to the [first automobile tutorial](blp.ipynb), but exhibits how to incorporate micro moments into estimation.\n",
-    "\n",
-    "\n",
-    "### Loading the Data\n",
+    "## Loading Data\n",
     "\n",
     "We'll use [pandas](https://pandas.pydata.org/) to load two sets of data:\n",
     "\n",
@@ -476,7 +471,7 @@
    "id": "a04056a2",
    "metadata": {},
    "source": [
-    "### Setting up the Problem\n",
+    "## Setting up the Problem\n",
     "\n",
     "The problem configuration is based on that of the first automobile problem. It is very similar, with both demand and supply sides, although with a few more product characteristics.\n",
     "\n",
@@ -614,7 +609,7 @@
    "id": "fde26d61",
    "metadata": {},
    "source": [
-    "### Setting up the Micro Moments"
+    "## Setting up Micro Moments"
    ]
   },
   {
@@ -1004,7 +999,7 @@
    "id": "c8443449",
    "metadata": {},
    "source": [
-    "### Solving the Problem\n",
+    "## Solving the Problem\n",
     "\n",
     "Like for the first automobile problem, here will will just use the publisehd estimates for $\\Sigma$ and $\\Pi$ starting values."
    ]
@@ -1177,7 +1172,7 @@
    "id": "8c5db574",
    "metadata": {},
    "source": [
-    "### Running the Main Counterfactual\n",
+    "## Running the Main Counterfactual\n",
     "\n",
     "One result that is very similar is the paper's headline number: a \\$367.29 million compensating variation from a counterfactual that removes the minivan in 1984. Using our estimates, we get a very similar number.\n",
     "\n",