AustinRochford/titanic_large_set.ipynb

## titanic_large_set.ipynb
{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "563a7d15",
   "metadata": {},
   "source": [
    "Lego recently announced the release of an extreme large (more than 9,000 pieces, roughly 4.5 ft long) [model of the Titanic](https://www.lego.com/en-us/product/lego-titanic-10294) (10294) priced at $629.99.\n",
    "\n",
    "<center><blockquote class=\"twitter-tweet\"><p lang=\"en\" dir=\"ltr\">9,090 pieces. 1.3 meters long (4 ft. 5 in). One LEGO Titanic building project! <a href=\"https://t.co/guhn2isu17\">https://t.co/guhn2isu17</a> <a href=\"https://t.co/jszY6C4MtC\">pic.twitter.com/jszY6C4MtC</a></p>&mdash; LEGO (@LEGO_Group) <a href=\"https://twitter.com/LEGO_Group/status/1446113247342383110?ref_src=twsrc%5Etfw\">October 7, 2021</a></blockquote> <script async src=\"https://platform.twitter.com/widgets.js\" charset=\"utf-8\"></script></center>\n",
    "\n",
    "<center><img alt=\"Lego model of the Titanic (10294)\" src=\"https://www.lego.com/cdn/cs/set/assets/bltb94632aeb971eb91/10294.jpg?fit=bounds&format=jpg&quality=80&width=600&height=600&dpr=1\"></center>\n",
    "\n",
    "Readers of this blog will know that I have written a series of posts ([1](https://austinrochford.com/posts/2021-06-03-vader-meditation.html) [2](https://austinrochford.com/posts/2021-06-10-lego-pymc3.html) [3](https://austinrochford.com/posts/2021-07-20-ucs-gunship.html)) analyzing the factors that drive Lego prices, from the year a set was released to its piece count, theme, and subtheme.  The impending release of the Titanic offers an excellent opportunity to use the model from the [most recent post](https://austinrochford.com/posts/2021-07-20-ucs-gunship.html) in this series to examine the prices of the sets with the largest number of pieces.\n",
    "\n",
    "First we make the necessary Python imports and do a bit of housekeeping."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "95bd0753",
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "9d1f3363",
   "metadata": {},
   "outputs": [],
   "source": [
    "import datetime\n",
    "from functools import reduce"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "27f4a808",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "You are running the v4 development version of PyMC3 which currently still lacks key features. You probably want to use the stable v3 instead which you can either install via conda or find on the v3 GitHub branch: https://github.com/pymc-devs/pymc3/tree/v3\n"
     ]
    }
   ],
   "source": [
    "from aesara import shared, tensor as at\n",
    "import arviz as az\n",
    "from matplotlib import MatplotlibDeprecationWarning, pyplot as plt\n",
    "from matplotlib.ticker import StrMethodFormatter\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import pymc3 as pm\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "import seaborn as sns\n",
    "import xarray as xr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "d0aa649a",
   "metadata": {},
   "outputs": [],
   "source": [
    "FIGSIZE = (8, 6)\n",
    "\n",
    "plt.rcParams['figure.figsize'] = FIGSIZE\n",
    "sns.set(color_codes=True)\n",
    "\n",
    "dollar_formatter = StrMethodFormatter(\"${x:,.0f}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9141a630",
   "metadata": {},
   "source": [
    "## Load the Data\n",
    "\n",
    "We begin the real work by loading the data scraped from Brickset. See the [first post](https://austinrochford.com/posts/2021-06-03-vader-meditation.html) in this series for more background on the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "8adebf24",
   "metadata": {},
   "outputs": [],
   "source": [
    "DATA_URL = 'https://austinrochford.com/resources/lego/brickset_19800101_20211098.csv.gz'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "4363585d",
   "metadata": {},
   "outputs": [],
   "source": [
    "def to_datetime(year):\n",
    "    return np.datetime64(f\"{round(year)}-01-01\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "a9709ff3",
   "metadata": {},
   "outputs": [],
   "source": [
    "COLS = [\n",
    "    \"Year released\", \"Set number\",\n",
    "    \"Name\", \"Set type\", \"Theme\", \"Subtheme\",\n",
    "    \"Pieces\", \"RRP$\"\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "9541854a",
   "metadata": {},
   "outputs": [],
   "source": [
    "full_df = (pd.read_csv(DATA_URL, usecols=COLS)\n",
    "             .dropna(subset=set(COLS) - {\"Subtheme\"}))\n",
    "full_df[\"Year released\"] = full_df[\"Year released\"].apply(to_datetime)\n",
    "full_df = (full_df.sort_values([\"Year released\", \"Set number\"])\n",
    "                  .set_index(\"Set number\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "3d577843",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Name</th>\n",
       "      <th>Set type</th>\n",
       "      <th>Theme</th>\n",
       "      <th>Year released</th>\n",
       "      <th>Pieces</th>\n",
       "      <th>Subtheme</th>\n",
       "      <th>RRP$</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Set number</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1041-2</th>\n",
       "      <td>Educational Duplo Building Set</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Dacta</td>\n",
       "      <td>1980-01-01</td>\n",
       "      <td>68.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>36.50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1075-1</th>\n",
       "      <td>LEGO People Supplementary Set</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Dacta</td>\n",
       "      <td>1980-01-01</td>\n",
       "      <td>304.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>14.50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1101-1</th>\n",
       "      <td>Replacement 4.5V Motor</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Service Packs</td>\n",
       "      <td>1980-01-01</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>5.65</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1123-1</th>\n",
       "      <td>Ball and Socket Couplings &amp; One Articulated Joint</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Service Packs</td>\n",
       "      <td>1980-01-01</td>\n",
       "      <td>8.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>16.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1130-1</th>\n",
       "      <td>Plastic Folder for Building Instructions</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Service Packs</td>\n",
       "      <td>1980-01-01</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>14.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                         Name Set type  \\\n",
       "Set number                                                               \n",
       "1041-2                         Educational Duplo Building Set   Normal   \n",
       "1075-1                          LEGO People Supplementary Set   Normal   \n",
       "1101-1                                 Replacement 4.5V Motor   Normal   \n",
       "1123-1      Ball and Socket Couplings & One Articulated Joint   Normal   \n",
       "1130-1               Plastic Folder for Building Instructions   Normal   \n",
       "\n",
       "                    Theme Year released  Pieces Subtheme   RRP$  \n",
       "Set number                                                       \n",
       "1041-2              Dacta    1980-01-01    68.0      NaN  36.50  \n",
       "1075-1              Dacta    1980-01-01   304.0      NaN  14.50  \n",
       "1101-1      Service Packs    1980-01-01     1.0      NaN   5.65  \n",
       "1123-1      Service Packs    1980-01-01     8.0      NaN  16.00  \n",
       "1130-1      Service Packs    1980-01-01     1.0      NaN  14.00  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "full_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "4be1d65e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Name</th>\n",
       "      <th>Set type</th>\n",
       "      <th>Theme</th>\n",
       "      <th>Year released</th>\n",
       "      <th>Pieces</th>\n",
       "      <th>Subtheme</th>\n",
       "      <th>RRP$</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Set number</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>80025-1</th>\n",
       "      <td>Sandy's Power Loader Mech</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Monkie Kid</td>\n",
       "      <td>2021-01-01</td>\n",
       "      <td>520.0</td>\n",
       "      <td>Season 2</td>\n",
       "      <td>54.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>80026-1</th>\n",
       "      <td>Pigsy's Noodle Tank</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Monkie Kid</td>\n",
       "      <td>2021-01-01</td>\n",
       "      <td>662.0</td>\n",
       "      <td>Season 2</td>\n",
       "      <td>59.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>80028-1</th>\n",
       "      <td>The Bone Demon</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Monkie Kid</td>\n",
       "      <td>2021-01-01</td>\n",
       "      <td>1375.0</td>\n",
       "      <td>Season 2</td>\n",
       "      <td>119.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>80106-1</th>\n",
       "      <td>Story of Nian</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Seasonal</td>\n",
       "      <td>2021-01-01</td>\n",
       "      <td>1067.0</td>\n",
       "      <td>Chinese Traditional Festivals</td>\n",
       "      <td>79.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>80107-1</th>\n",
       "      <td>Spring Lantern Festival</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Seasonal</td>\n",
       "      <td>2021-01-01</td>\n",
       "      <td>1793.0</td>\n",
       "      <td>Chinese Traditional Festivals</td>\n",
       "      <td>119.99</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 Name Set type       Theme Year released  \\\n",
       "Set number                                                                 \n",
       "80025-1     Sandy's Power Loader Mech   Normal  Monkie Kid    2021-01-01   \n",
       "80026-1           Pigsy's Noodle Tank   Normal  Monkie Kid    2021-01-01   \n",
       "80028-1                The Bone Demon   Normal  Monkie Kid    2021-01-01   \n",
       "80106-1                 Story of Nian   Normal    Seasonal    2021-01-01   \n",
       "80107-1       Spring Lantern Festival   Normal    Seasonal    2021-01-01   \n",
       "\n",
       "            Pieces                       Subtheme    RRP$  \n",
       "Set number                                                 \n",
       "80025-1      520.0                       Season 2   54.99  \n",
       "80026-1      662.0                       Season 2   59.99  \n",
       "80028-1     1375.0                       Season 2  119.99  \n",
       "80106-1     1067.0  Chinese Traditional Festivals   79.99  \n",
       "80107-1     1793.0  Chinese Traditional Festivals  119.99  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "full_df.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "cfc002d9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Pieces</th>\n",
       "      <th>RRP$</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>8636.000000</td>\n",
       "      <td>8636.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>265.029180</td>\n",
       "      <td>31.741934</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>500.511219</td>\n",
       "      <td>46.129177</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>32.000000</td>\n",
       "      <td>7.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>100.000000</td>\n",
       "      <td>18.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>305.000000</td>\n",
       "      <td>39.990000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>11695.000000</td>\n",
       "      <td>799.990000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Pieces         RRP$\n",
       "count   8636.000000  8636.000000\n",
       "mean     265.029180    31.741934\n",
       "std      500.511219    46.129177\n",
       "min        0.000000     0.000000\n",
       "25%       32.000000     7.000000\n",
       "50%      100.000000    18.000000\n",
       "75%      305.000000    39.990000\n",
       "max    11695.000000   799.990000"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "full_df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1168f0a8",
   "metadata": {},
   "source": [
    "We now add a column `RRP2021`, which is `RRP$` adjusted to 2021 dollars."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "68b82097",
   "metadata": {},
   "outputs": [],
   "source": [
    "CPI_URL = 'https://austinrochford.com/resources/lego/CPIAUCNS202100401.csv'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "7947fa09",
   "metadata": {},
   "outputs": [],
   "source": [
    "years = pd.date_range('1979-01-01', '2021-01-01', freq='Y') \\\n",
    "            + datetime.timedelta(days=1)\n",
    "cpi_df = (pd.read_csv(CPI_URL, index_col=\"DATE\", parse_dates=[\"DATE\"])\n",
    "            .loc[years])\n",
    "cpi_df[\"to2021\"] = cpi_df.loc[\"2021-01-01\"] / cpi_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "6dad223c",
   "metadata": {},
   "outputs": [],
   "source": [
    "full_df[\"RRP2021\"] = (pd.merge(full_df, cpi_df,\n",
    "                               left_on=[\"Year released\"],\n",
    "                               right_index=True)\n",
    "                        .apply(lambda df: df[\"RRP$\"] * df[\"to2021\"],\n",
    "                               axis=1))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "de332aea",
   "metadata": {},
   "source": [
    "Based on the exploratory data analysis in the first post in this series, we filter `full_df` down to approximately 6,400 sets to be included in our analysis."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "dcf69465",
   "metadata": {},
   "outputs": [],
   "source": [
    "FILTERS = [\n",
    "    full_df[\"Set type\"] == \"Normal\",\n",
    "    full_df[\"Pieces\"] > 10,\n",
    "    full_df[\"Theme\"] != \"Duplo\",\n",
    "    full_df[\"Theme\"] != \"Service Packs\",\n",
    "    full_df[\"Theme\"] != \"Bulk Bricks\",\n",
    "    full_df[\"RRP2021\"] > 0\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "598613a4",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = full_df[reduce(np.logical_and, FILTERS)].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "837138a0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Name</th>\n",
       "      <th>Set type</th>\n",
       "      <th>Theme</th>\n",
       "      <th>Year released</th>\n",
       "      <th>Pieces</th>\n",
       "      <th>Subtheme</th>\n",
       "      <th>RRP$</th>\n",
       "      <th>RRP2021</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Set number</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1041-2</th>\n",
       "      <td>Educational Duplo Building Set</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Dacta</td>\n",
       "      <td>1980-01-01</td>\n",
       "      <td>68.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>36.50</td>\n",
       "      <td>122.721632</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1075-1</th>\n",
       "      <td>LEGO People Supplementary Set</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Dacta</td>\n",
       "      <td>1980-01-01</td>\n",
       "      <td>304.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>14.50</td>\n",
       "      <td>48.752429</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5233-1</th>\n",
       "      <td>Bedroom</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Homemaker</td>\n",
       "      <td>1980-01-01</td>\n",
       "      <td>26.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.50</td>\n",
       "      <td>15.130064</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6305-1</th>\n",
       "      <td>Trees and Flowers</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Town</td>\n",
       "      <td>1980-01-01</td>\n",
       "      <td>12.0</td>\n",
       "      <td>Accessories</td>\n",
       "      <td>3.75</td>\n",
       "      <td>12.608387</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6306-1</th>\n",
       "      <td>Road Signs</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Town</td>\n",
       "      <td>1980-01-01</td>\n",
       "      <td>12.0</td>\n",
       "      <td>Accessories</td>\n",
       "      <td>2.50</td>\n",
       "      <td>8.405591</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                      Name Set type      Theme Year released  \\\n",
       "Set number                                                                     \n",
       "1041-2      Educational Duplo Building Set   Normal      Dacta    1980-01-01   \n",
       "1075-1       LEGO People Supplementary Set   Normal      Dacta    1980-01-01   \n",
       "5233-1                             Bedroom   Normal  Homemaker    1980-01-01   \n",
       "6305-1                   Trees and Flowers   Normal       Town    1980-01-01   \n",
       "6306-1                          Road Signs   Normal       Town    1980-01-01   \n",
       "\n",
       "            Pieces     Subtheme   RRP$     RRP2021  \n",
       "Set number                                          \n",
       "1041-2        68.0          NaN  36.50  122.721632  \n",
       "1075-1       304.0          NaN  14.50   48.752429  \n",
       "5233-1        26.0          NaN   4.50   15.130064  \n",
       "6305-1        12.0  Accessories   3.75   12.608387  \n",
       "6306-1        12.0  Accessories   2.50    8.405591  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "33baebec",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Name</th>\n",
       "      <th>Set type</th>\n",
       "      <th>Theme</th>\n",
       "      <th>Year released</th>\n",
       "      <th>Pieces</th>\n",
       "      <th>Subtheme</th>\n",
       "      <th>RRP$</th>\n",
       "      <th>RRP2021</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Set number</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>80025-1</th>\n",
       "      <td>Sandy's Power Loader Mech</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Monkie Kid</td>\n",
       "      <td>2021-01-01</td>\n",
       "      <td>520.0</td>\n",
       "      <td>Season 2</td>\n",
       "      <td>54.99</td>\n",
       "      <td>54.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>80026-1</th>\n",
       "      <td>Pigsy's Noodle Tank</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Monkie Kid</td>\n",
       "      <td>2021-01-01</td>\n",
       "      <td>662.0</td>\n",
       "      <td>Season 2</td>\n",
       "      <td>59.99</td>\n",
       "      <td>59.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>80028-1</th>\n",
       "      <td>The Bone Demon</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Monkie Kid</td>\n",
       "      <td>2021-01-01</td>\n",
       "      <td>1375.0</td>\n",
       "      <td>Season 2</td>\n",
       "      <td>119.99</td>\n",
       "      <td>119.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>80106-1</th>\n",
       "      <td>Story of Nian</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Seasonal</td>\n",
       "      <td>2021-01-01</td>\n",
       "      <td>1067.0</td>\n",
       "      <td>Chinese Traditional Festivals</td>\n",
       "      <td>79.99</td>\n",
       "      <td>79.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>80107-1</th>\n",
       "      <td>Spring Lantern Festival</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Seasonal</td>\n",
       "      <td>2021-01-01</td>\n",
       "      <td>1793.0</td>\n",
       "      <td>Chinese Traditional Festivals</td>\n",
       "      <td>119.99</td>\n",
       "      <td>119.99</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 Name Set type       Theme Year released  \\\n",
       "Set number                                                                 \n",
       "80025-1     Sandy's Power Loader Mech   Normal  Monkie Kid    2021-01-01   \n",
       "80026-1           Pigsy's Noodle Tank   Normal  Monkie Kid    2021-01-01   \n",
       "80028-1                The Bone Demon   Normal  Monkie Kid    2021-01-01   \n",
       "80106-1                 Story of Nian   Normal    Seasonal    2021-01-01   \n",
       "80107-1       Spring Lantern Festival   Normal    Seasonal    2021-01-01   \n",
       "\n",
       "            Pieces                       Subtheme    RRP$  RRP2021  \n",
       "Set number                                                          \n",
       "80025-1      520.0                       Season 2   54.99    54.99  \n",
       "80026-1      662.0                       Season 2   59.99    59.99  \n",
       "80028-1     1375.0                       Season 2  119.99   119.99  \n",
       "80106-1     1067.0  Chinese Traditional Festivals   79.99    79.99  \n",
       "80107-1     1793.0  Chinese Traditional Festivals  119.99   119.99  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "6e8a749f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Pieces</th>\n",
       "      <th>RRP$</th>\n",
       "      <th>RRP2021</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>6423.000000</td>\n",
       "      <td>6423.000000</td>\n",
       "      <td>6423.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>345.121283</td>\n",
       "      <td>37.652064</td>\n",
       "      <td>46.267159</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>556.907975</td>\n",
       "      <td>50.917343</td>\n",
       "      <td>59.812083</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>11.000000</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>0.971220</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>69.000000</td>\n",
       "      <td>9.990000</td>\n",
       "      <td>11.896044</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>181.000000</td>\n",
       "      <td>19.990000</td>\n",
       "      <td>27.420158</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>404.000000</td>\n",
       "      <td>49.990000</td>\n",
       "      <td>56.497192</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>11695.000000</td>\n",
       "      <td>799.990000</td>\n",
       "      <td>897.373477</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Pieces         RRP$      RRP2021\n",
       "count   6423.000000  6423.000000  6423.000000\n",
       "mean     345.121283    37.652064    46.267159\n",
       "std      556.907975    50.917343    59.812083\n",
       "min       11.000000     0.600000     0.971220\n",
       "25%       69.000000     9.990000    11.896044\n",
       "50%      181.000000    19.990000    27.420158\n",
       "75%      404.000000    49.990000    56.497192\n",
       "max    11695.000000   799.990000   897.373477"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0d088407",
   "metadata": {},
   "source": [
    "## Modeling\n",
    "\n",
    "We will reuse the final model from the [previous post](https://austinrochford.com/posts/2021-07-20-ucs-gunship.html).  It includes a time-varying component along with random slopes and intercepts that vary according to the the theme and subtheme of the set in question.\n",
    "\n",
    "First we create our feature vector (standardized log piece count) and our target vector (log RRP in 2021 dollars)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "835bca3c",
   "metadata": {},
   "outputs": [],
   "source": [
    "pieces = df[\"Pieces\"].values\n",
    "log_pieces = np.log(df[\"Pieces\"].values)\n",
    "\n",
    "scaler = StandardScaler().fit(log_pieces[:, np.newaxis])\n",
    "\n",
    "def scale_log_pieces(log_pieces, scaler=scaler):\n",
    "    return scaler.transform(log_pieces[:, np.newaxis])[:, 0]\n",
    "\n",
    "std_log_pieces = scale_log_pieces(log_pieces)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "1d5eefa6",
   "metadata": {},
   "outputs": [],
   "source": [
    "rrp2021 = df[\"RRP2021\"].values\n",
    "log_rrp2021 = np.log(rrp2021)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6e4774b2",
   "metadata": {},
   "source": [
    "Next we build indicator variables for the themes and subthemes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "6ab0c435",
   "metadata": {},
   "outputs": [],
   "source": [
    "theme_id, theme_map = df[\"Theme\"].factorize(sort=True)\n",
    "\n",
    "theme_mean_std_log_pieces = (pd.Series(std_log_pieces, index=df.index)\n",
    "                               .groupby(df[\"Theme\"])\n",
    "                               .mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "614d5712",
   "metadata": {},
   "outputs": [],
   "source": [
    "sub_id, sub_map = df[\"Subtheme\"].factorize(sort=True)\n",
    "n_sub = sub_map.size\n",
    "\n",
    "sub_isnull = sub_id == -1\n",
    "\n",
    "sub_mean_std_log_pieces = (pd.Series(std_log_pieces, index=df.index)\n",
    "                             .groupby(df[\"Subtheme\"])\n",
    "                             .mean())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1f8fb5ab",
   "metadata": {},
   "source": [
    "Finally, we transform the year each set was released into the number of years since 1980, when our data set begins."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "73024138",
   "metadata": {},
   "outputs": [],
   "source": [
    "year = (df[\"Year released\"]\n",
    "          .dt.year\n",
    "          .values)\n",
    "t = year - year.min()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9ded232b",
   "metadata": {},
   "source": [
    "We now recreate the model as in the previous post."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "9d1b163e",
   "metadata": {},
   "outputs": [],
   "source": [
    "def hierarchical_normal(name, *, dims, μ=None):\n",
    "    if μ is None:\n",
    "        μ = pm.Normal(f\"μ_{name}\", 0., 2.5)\n",
    "\n",
    "    Δ = pm.Normal(f\"Δ_{name}\", 0., 1., dims=dims)\n",
    "    σ = pm.HalfNormal(f\"σ_{name}\", 2.5)\n",
    "    \n",
    "    return pm.Deterministic(name, μ + Δ * σ, dims=dims)\n",
    "\n",
    "def hierarchical_normal_with_mean(name, x_mean, *, dims,\n",
    "                                  μ=None, mean_name=\"γ\"):\n",
    "    mean_coef = pm.Normal(f\"{mean_name}_{name}\", 0., 2.5)\n",
    "        \n",
    "    return pm.Deterministic(\n",
    "        name,\n",
    "        hierarchical_normal(f\"_{name}\", dims=dims, μ=μ) \\\n",
    "            + mean_coef * x_mean,\n",
    "        dims=dims\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "29d468b3",
   "metadata": {},
   "outputs": [],
   "source": [
    "def gaussian_random_walk(name, *, dims, innov_scale=1.):\n",
    "    Δ = pm.Normal(f\"Δ_{name}\", 0., innov_scale,  dims=dims)\n",
    "\n",
    "    return pm.Deterministic(name, Δ.cumsum(), dims=dims)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "7c961a15",
   "metadata": {},
   "outputs": [],
   "source": [
    "coords = {\n",
    "    \"set\": df.index.values,\n",
    "    \"sub\": sub_map,\n",
    "    \"theme\": theme_map,\n",
    "    \"year\": np.unique(year)\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e37507b1",
   "metadata": {},
   "source": [
    "First we specify the components of the intercept."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "78ec36d8",
   "metadata": {},
   "outputs": [],
   "source": [
    "with pm.Model(coords=coords) as model:\n",
    "    β0_0 = pm.Normal(\"β0_0\", 0., 2.5)\n",
    "    β0_t = gaussian_random_walk(\"β0_t\", dims=\"year\",\n",
    "                                innov_scale=0.1)\n",
    "    \n",
    "    β0_theme = hierarchical_normal_with_mean(\n",
    "        \"β0_theme\",\n",
    "        theme_mean_std_log_pieces.values,\n",
    "        dims=\"theme\", μ=0.\n",
    "    )\n",
    "    \n",
    "    β0_sub_null = pm.Normal(\"β0_sub_null\", 0., 2.5)\n",
    "    β0_sub_nn = hierarchical_normal_with_mean(\n",
    "        \"β0_sub_nn\",\n",
    "        sub_mean_std_log_pieces.values,\n",
    "        dims=\"sub\", μ=0., mean_name=\"λ\"\n",
    "    )\n",
    "    β0_sub_i = at.switch(\n",
    "        sub_isnull,\n",
    "        β0_sub_null,\n",
    "        β0_sub_nn[at.clip(sub_id, 0, n_sub)]\n",
    "    )\n",
    "    \n",
    "    β0_i = β0_0 + β0_t[t] + β0_theme[theme_id] + β0_sub_i"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "05106555",
   "metadata": {},
   "source": [
    "We specify the components of the slope for the (standardized log) number of pieces in the set analagously."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "9b8c2a91",
   "metadata": {},
   "outputs": [],
   "source": [
    "with model:\n",
    "    β_pieces_0 = pm.Normal(\"β_pieces_0\", 0., 2.5)\n",
    "    \n",
    "    β_pieces_theme = hierarchical_normal_with_mean(\n",
    "        \"β_pieces_theme\",\n",
    "        theme_mean_std_log_pieces.values,\n",
    "        dims=\"theme\", μ=0.\n",
    "    )\n",
    "    \n",
    "    β_pieces_sub_null = pm.Normal(\"β_pieces_sub_null\", 0., 2.5)\n",
    "    β_pieces_sub_nn = hierarchical_normal_with_mean(\n",
    "        \"β_pieces_sub_nn\",\n",
    "        sub_mean_std_log_pieces.values,\n",
    "        dims=\"sub\", μ=0., mean_name=\"λ\"\n",
    "    )\n",
    "    β_pieces_sub_i = at.switch(\n",
    "        sub_isnull,\n",
    "        β_pieces_sub_null,\n",
    "        β_pieces_sub_nn[at.clip(sub_id, 0, n_sub)]\n",
    "    )\n",
    "\n",
    "    β_pieces_i = β_pieces_0 + β_pieces_theme[theme_id] + β_pieces_sub_i"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "845423b5",
   "metadata": {},
   "source": [
    "Finally we specify the observational likelihood."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "ebeebc67",
   "metadata": {},
   "outputs": [],
   "source": [
    "with model:\n",
    "    σ = pm.HalfNormal(\"σ\", 5.)\n",
    "    μ = β0_i + β_pieces_i * std_log_pieces - 0.5 * σ**2\n",
    "    \n",
    "    obs = pm.Normal(\"obs\", μ, σ, dims=\"set\",\n",
    "                    observed=log_rrp2021)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6c5d5948",
   "metadata": {},
   "source": [
    "We are now ready to sample from the posterior distribution of this model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "7cb3d3a0",
   "metadata": {},
   "outputs": [],
   "source": [
    "CHAINS = 3\n",
    "SEED = 123456789\n",
    "\n",
    "SAMPLE_KWARGS = {\n",
    "    'cores': CHAINS,\n",
    "    'random_seed': [SEED + i for i in range(CHAINS)],\n",
    "    'return_inferencedata': True\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "db9d34e5",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Auto-assigning NUTS sampler...\n",
      "Initializing NUTS using jitter+adapt_diag...\n",
      "Multiprocess sampling (3 chains in 3 jobs)\n",
      "NUTS: [β0_0, Δ_β0_t, γ_β0_theme, Δ__β0_theme, σ__β0_theme, β0_sub_null, λ_β0_sub_nn, Δ__β0_sub_nn, σ__β0_sub_nn, β_pieces_0, γ_β_pieces_theme, Δ__β_pieces_theme, σ__β_pieces_theme, β_pieces_sub_null, λ_β_pieces_sub_nn, Δ__β_pieces_sub_nn, σ__β_pieces_sub_nn, σ]\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "    <div>\n",
       "        <style>\n",
       "            /* Turns off some styling */\n",
       "            progress {\n",
       "                /* gets rid of default border in Firefox and Opera. */\n",
       "                border: none;\n",
       "                /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "                background-size: auto;\n",
       "            }\n",
       "            .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "                background: #F44336;\n",
       "            }\n",
       "        </style>\n",
       "      <progress value='6000' class='' max='6000' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
       "      100.00% [6000/6000 17:15<00:00 Sampling 3 chains, 0 divergences]\n",
       "    </div>\n",
       "    "
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 3 chains for 1_000 tune and 1_000 draw iterations (3_000 + 3_000 draws total) took 1036 seconds.\n",
      "The number of effective samples is smaller than 10% for some parameters.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "    <div>\n",
       "        <style>\n",
       "            /* Turns off some styling */\n",
       "            progress {\n",
       "                /* gets rid of default border in Firefox and Opera. */\n",
       "                border: none;\n",
       "                /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "                background-size: auto;\n",
       "            }\n",
       "            .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "                background: #F44336;\n",
       "            }\n",
       "        </style>\n",
       "      <progress value='3000' class='' max='3000' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
       "      100.00% [3000/3000 00:02<00:00]\n",
       "    </div>\n",
       "    "
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "with model:\n",
    "    trace = pm.sample(**SAMPLE_KWARGS)\n",
    "    pp_trace = pm.to_inference_data(\n",
    "        posterior_predictive=pm.sample_posterior_predictive(trace)\n",
    "    )\n",
    "\n",
    "trace.extend(pp_trace)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9049ec23",
   "metadata": {},
   "source": [
    "Standard sampling diangostics show no cause for concern."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "266288e5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "az.plot_energy(trace);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "598a44ca",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<xarray.DataArray ()>\n",
      "array(1.01564337)\n"
     ]
    }
   ],
   "source": [
    "print(az.rhat(trace)\n",
    "        .max()\n",
    "        .to_array()\n",
    "        .max())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "36149e13",
   "metadata": {},
   "source": [
    "## Analysis\n",
    "\n",
    "Now that we have sampled from the posterior distribution of the model, we are finally prepared to analyze and interpret the pricing of the sets with the largest number of pieces.  First we select the twenty sets with the largest piece count for closer inspection."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "124bfed2",
   "metadata": {},
   "outputs": [],
   "source": [
    "N_TOP = 20\n",
    "\n",
    "top_by_pieces = df.nlargest(N_TOP, \"Pieces\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c8e05401",
   "metadata": {},
   "source": [
    "We examine the posterior predictive distributions of each of these set's price compared to its actual recommended retail price below."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "518e8b1c",
   "metadata": {},
   "outputs": [],
   "source": [
    "def make_set_label(row):\n",
    "    name = row[\"Name\"]\n",
    "    set_number, _ = row.name.split('-', 1)\n",
    "    pieces = int(row[\"Pieces\"])\n",
    "    \n",
    "    return f\"{name} ({set_number})\\n{pieces:,} pieces\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "38e4219d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x864 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(FIGSIZE[0], 2 * FIGSIZE[1]))\n",
    "\n",
    "az.plot_forest(\n",
    "    trace.posterior_predictive,\n",
    "    var_names=[\"obs\"],\n",
    "    coords={\"set\": top_by_pieces.index},\n",
    "    transform=np.exp,\n",
    "    ax=ax\n",
    ");\n",
    "ax.scatter(\n",
    "    top_by_pieces[\"RRP2021\"][::-1],\n",
    "    ax.get_yticks(),\n",
    "    c='k', label=\"Actual\", zorder=5\n",
    ");\n",
    "\n",
    "ax.set_yticklabels(\n",
    "    top_by_pieces.apply(make_set_label, axis=1)[::-1]\n",
    ");\n",
    "\n",
    "ax.xaxis.set_major_formatter(dollar_formatter);\n",
    "ax.set_xlabel(\"Posterior predicted RRP (2021 $)\");\n",
    "\n",
    "ax.legend();\n",
    "ax.set_title(\"Sets with the most pieces\");"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a5f4d4fe",
   "metadata": {},
   "source": [
    "The first time I drew this plot I was a bit surprised at how well the model predicts the prices for these large sets.  I expected to see larger deviations for sets with an extreme number of pieces (more than about 3,000), so this is a pleasant surprise.  It is worth nothing that while the model's predictions are accurate for most of these large sets, there are a few sets where the actual price are further from the posterior expected value.  These sets are worth focusing on."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "136fe017",
   "metadata": {},
   "outputs": [],
   "source": [
    "FOCUS_SETS = [\n",
    "    \"31203-1\",\n",
    "    \"10294-1\",\n",
    "    \"71043-1\",\n",
    "    \"2000409-1\",\n",
    "    \"75252-1\",\n",
    "    \"10214-1\"\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "0be5bd60",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "\n",
    "az.plot_forest(\n",
    "    trace.posterior_predictive,\n",
    "    var_names=[\"obs\"],\n",
    "    coords={\"set\": FOCUS_SETS},\n",
    "    transform=np.exp,\n",
    "    ax=ax\n",
    ");\n",
    "ax.scatter(\n",
    "    df[\"RRP2021\"]\n",
    "      .loc[FOCUS_SETS]\n",
    "      [::-1],\n",
    "    ax.get_yticks(),\n",
    "    c='k', label=\"Actual\", zorder=5\n",
    ");\n",
    "\n",
    "ax.set_yticklabels(\n",
    "    df.loc[FOCUS_SETS]\n",
    "      .apply(make_set_label, axis=1)\n",
    "      [::-1]\n",
    ");\n",
    "\n",
    "ax.xaxis.set_major_formatter(dollar_formatter);\n",
    "ax.set_xlabel(\"Posterior predicted RRP (2021 $)\");\n",
    "\n",
    "ax.legend();\n",
    "ax.set_title(None);"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ddd8e15e",
   "metadata": {},
   "source": [
    "Sets that are priced significantly lower than their posterior expected value tend to have lots of similar small and/or standard pieces.\n",
    "\n",
    "* The [World Map (31203)](https://www.lego.com/en-gb/product/world-map-31203) contains thousands of small dots.\n",
    "\n",
    "<center><img alt=\"Lego World Map (31203)\" src=\"https://www.lego.com/cdn/cs/set/assets/blt6e4605758d75ce41/31203.jpg?fit=bounds&format=jpg&quality=80&width=600&height=600&dpr=1\"></center>\n",
    "\n",
    "* The [Window Exploration Bag (2000409)](https://www.lego.com/en-us/product/window-exploration-bag-2000409) contains thousands of fairly standard bricks.  (I had never heard of Lego's [\"Serious Play\"](https://www.lego.com/en-us/seriousplay) method before, and it is interesting and may merit a post of its own.)\n",
    "\n",
    "<center><img alt=\"Lego Window Exploration Bag (2000409)\" src=\"https://www.lego.com/cdn/cs/set/assets/bltf952785f84bbefa6/2000409.jpg?fit=bounds&format=jpg&quality=80&width=600&height=600&dpr=1\"></center>\n",
    "\n",
    "* [Tower Bridge (10214)](https://www.lego.com/en-us/product/tower-bridge-10214) conforms to this explanation slightly less well, but there are still large repetitive portions (turrets, bridge suspension cables) that require many fairly standard pieces.\n",
    "\n",
    "<center><img alt=\"Lego Tower Bridge (10214)\" src=\"https://www.lego.com/cdn/cs/set/assets/bltc4f19e085e17bcea/10214.jpg?fit=bounds&format=jpg&quality=80&width=600&height=600&dpr=1\"></center>\n",
    "\n",
    "In light of these similarities for large sets that are underpriced relative to our predictions, it makes sense to consider including information about the pieces included in the set (and the quantity of each piece that is included) in a future model.  I have not scraped this data about the piece composition of each set yet, but it is available from [Brickset](https://brickset.com/) an probably also from [BrickLink](https://bricklink.com).\n",
    "\n",
    "Sets that are priced significantly higher than their posterior expected value include the Titanic (10294), [Hogwarts Castle (71043)](https://www.lego.com/en-us/product/hogwarts-castle-71043) and the [Ultimate Collector Series Imperial Star Destroyer (75252)](https://www.lego.com/en-us/product/imperial-star-destroyer-75252).  Each of these sets is iconic either on its own (the Titanic) or within the associated media franchies (Harry Potter and Star Wars).  It seems reasonable that Lego may be able to charge a premium on such sets.  Note that this interpretation does not hold true for all sets, as the [Ultimate Collector Series Millenium Falcon (75192)](https://www.lego.com/en-us/product/millennium-falcon-75192) is perhaps even more iconic than the Imperial Star Destroyer.  In this instance, Lego may believe that the recommended retail price of $799.99 is near the maximum that the market will reasonably bear for a Lego set.\n",
    "\n",
    "We can examine the posterior predictive distribution of some individual sets in more detail."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "efe9f1cd",
   "metadata": {},
   "outputs": [],
   "source": [
    "def format_posterior_artist(artist, formatter):\n",
    "    text = artist.get_text()\n",
    "    x, _ = artist.get_position()\n",
    "\n",
    "    if text.startswith(\" \") or text.endswith(\" \"):\n",
    "        fmtd_text = formatter(x)\n",
    "        artist.set_text(\n",
    "            \" \" + fmtd_text if text.startswith(\" \") else fmtd_text + \" \"\n",
    "        )\n",
    "    elif \"=\" in text:\n",
    "        before, _ = text.split(\"=\")\n",
    "        artist.set_text(\"=\".join((before, formatter(x))))\n",
    "    elif \"<\" in text:\n",
    "        below, ref_val_str, above = text.split(\"<\")\n",
    "        artist.set_text(\"<\".join((\n",
    "            below,\n",
    "            \" \" + formatter(float(ref_val_str)) + \" \",\n",
    "            above\n",
    "        )))\n",
    "\n",
    "def format_posterior_text(formatter, ax=None):\n",
    "    if ax is None:\n",
    "        ax = plt.gca()\n",
    "    \n",
    "    artists = [artist for artist in ax.get_children() if isinstance(artist, plt.Text)]\n",
    "    \n",
    "    for artist in artists:\n",
    "        format_posterior_artist(artist, formatter)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "1bd4b768",
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_set_posterior(set_number, df, ax=None):\n",
    "    if ax is None:\n",
    "        fig, ax = plt.subplots()\n",
    "    \n",
    "    az.plot_posterior(\n",
    "        trace, var_names=[\"obs\"],\n",
    "        group=\"posterior_predictive\",\n",
    "        coords={\"set\": f\"{set_number}-1\"},\n",
    "        transform=np.exp,\n",
    "        ref_val=df[\"RRP2021\"].loc[f\"{set_number}-1\"],\n",
    "        ax=ax\n",
    "    )\n",
    "\n",
    "    format_posterior_text(dollar_formatter, ax=ax)\n",
    "\n",
    "    ax.set_xlabel(\"Posterior predicted RRP (2021 $)\")\n",
    "    ax.set_title(f\"{df['Name'].loc[set_number + '-1']} ({set_number})\")\n",
    "    \n",
    "    return ax"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5ff123f4",
   "metadata": {},
   "source": [
    "As seen above, the Titanic (10294) is a bit overpriced according to our model, but not unreasonably so."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "212983d7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "TITANIC_SET_NUMBER = \"10294\"\n",
    "plot_set_posterior(TITANIC_SET_NUMBER, df);"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "27e3942b",
   "metadata": {},
   "source": [
    "We see that the Millenium Falcon (75192) is priced right about in line with the models expectations."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "f5e12702",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "MF_SET_NUMBER = \"75192\"\n",
    "plot_set_posterior(MF_SET_NUMBER, df);"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e929347d",
   "metadata": {},
   "source": [
    "I hope this analysis has been as interesting to you as it has to me. I am pleasantly surprised at how well the model predicts set prices even when the number of pieces is extremely large.  While this analysis has not sold me on the Titanic, it adds to my desire to get the Ultimate Collector Series Millenium Falcon and Imperial Star Destroy some day. (Unfortunately?) I have to finish my half-built [Super Star Destroyer (10221)](https://brickset.com/sets/10221-1/Super-Star-Destroyer) first."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "d8bfbf15",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Last updated: Tue Oct 12 2021\n",
      "\n",
      "Python implementation: CPython\n",
      "Python version       : 3.7.10\n",
      "IPython version      : 7.25.0\n",
      "\n",
      "arviz     : 0.11.2\n",
      "seaborn   : 0.11.1\n",
      "matplotlib: 3.4.2\n",
      "aesara    : 2.1.3\n",
      "xarray    : 0.19.0\n",
      "pymc3     : 4.0\n",
      "numpy     : 1.19.5\n",
      "pandas    : 1.3.0\n",
      "\n"
     ]
    }
   ],
   "source": [
    "%load_ext watermark\n",
    "%watermark -n -u -v -iv"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}