{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "408a6524",
   "metadata": {},
   "source": [
    "# Information Value analysis with the **vivainsights** Python library\n",
    "\n",
    "This notebook provides a demo on the Information Value (IV) functions for the **vivainsights** package. For more information about the package, please see:\n",
    "- [Documentation](https://microsoft.github.io/vivainsights-py/)\n",
    "- [GitHub Page](https://github.com/microsoft/vivainsights-py/)\n",
    "\n",
    "In this notebook, we will demo how to create analysis and visualizations with the IV and plot-WOE queries from Viva Insights.\n",
    "\n",
    "## Background\n",
    "\n",
    "Information Value (IV) is a powerful methodology that provides a measure of the predictive power of an individual independent variable in relation to the dependent variable. In the context of Viva Insights, independent variables could be a collaboration metric (e.g. Emails sent, 1:1 meeting time with managers), whereas a dependent variable could be a categorical variable indicating whether a person is engaged, a top performer, or at risk of attrition - likely provided through a survey. \n",
    "\n",
    "IV quantifies the amount of information a variable provides about the outcome. It is based on the following logic: a variable that is highly informative of the outcome will have different distributions of values for different outcome classes. For example, if we are predicting employee engagement, a variable like collaboration hours might have a different distribution for the engaged and non-engaged classes, indicating that it is informative of the outcome.\n",
    "\n",
    "The IV is calculated for each potential predictor variable, and the variables are then ranked based on their IVs. This allows for the selection of the most predictive variables for use in the model. The IV methodology solves the problem of selecting the most predictive variables for a predictive model. By ranking variables based on their IVs, it allows for the selection of variables that are most informative of the outcome, improving the predictive power of the model. It also helps in identifying and excluding variables that are not predictive of the outcome, which can improve model performance and interpretability.\n",
    "\n",
    "## Set up\n",
    "\n",
    "We start with loading the **vivainsights** package, and loading the default person query dataset with `load_pq_data()`:\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "bf518f05",
   "metadata": {},
   "outputs": [],
   "source": [
    "import vivainsights as vi\n",
    "import numpy as np\n",
    "\n",
    "# load in-built datasets\n",
    "pq_data = vi.load_pq_data() # load and assign in-built"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d9e05451",
   "metadata": {},
   "source": [
    "The following shows a preview of the Person Query demo dataset: "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "bbd364ce",
   "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>Unnamed: 0</th>\n",
       "      <th>PersonId</th>\n",
       "      <th>MetricDate</th>\n",
       "      <th>After_hours_call_hours</th>\n",
       "      <th>After_hours_chat_hours</th>\n",
       "      <th>After_hours_collaboration_hours</th>\n",
       "      <th>After_hours_email_hours</th>\n",
       "      <th>After_hours_meeting_hours</th>\n",
       "      <th>After_hours_scheduled_call_hours</th>\n",
       "      <th>After_hours_unscheduled_call_hours</th>\n",
       "      <th>...</th>\n",
       "      <th>Working_hours_meeting_hours</th>\n",
       "      <th>Working_hours_scheduled_call_hours</th>\n",
       "      <th>Working_hours_unscheduled_call_hours</th>\n",
       "      <th>LevelDesignation</th>\n",
       "      <th>Layer</th>\n",
       "      <th>SupervisorIndicator</th>\n",
       "      <th>Organization</th>\n",
       "      <th>FunctionType</th>\n",
       "      <th>WeekendDays</th>\n",
       "      <th>IsActive</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>a6afe34c-8524-32d3-a368-1517b29b68cd</td>\n",
       "      <td>2022-05-01</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>18.675938</td>\n",
       "      <td>0.722722</td>\n",
       "      <td>18.25</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>19.50</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Manager</td>\n",
       "      <td>3</td>\n",
       "      <td>Manager</td>\n",
       "      <td>Sales and Marketing</td>\n",
       "      <td>G_and_A</td>\n",
       "      <td>[SUNDAY, SATURDAY]</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>d6368140-9312-380b-bbc9-9a32bcef4b83</td>\n",
       "      <td>2022-05-01</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.827803</td>\n",
       "      <td>0.925556</td>\n",
       "      <td>4.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>8.75</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Support</td>\n",
       "      <td>3</td>\n",
       "      <td>Individual Contributor</td>\n",
       "      <td>Finance</td>\n",
       "      <td>Sales</td>\n",
       "      <td>[SUNDAY, SATURDAY]</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>60bf99b0-65fd-3c3f-94fb-8ceb451d59e7</td>\n",
       "      <td>2022-05-01</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.497806</td>\n",
       "      <td>0.812806</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>12.50</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Support</td>\n",
       "      <td>3</td>\n",
       "      <td>Individual Contributor</td>\n",
       "      <td>Product</td>\n",
       "      <td>IT</td>\n",
       "      <td>[SUNDAY, SATURDAY]</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>93fddd74-3667-392b-ba5a-92d855772cb0</td>\n",
       "      <td>2022-05-01</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>59.265892</td>\n",
       "      <td>2.283668</td>\n",
       "      <td>59.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>28.50</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Director</td>\n",
       "      <td>2</td>\n",
       "      <td>Manager+</td>\n",
       "      <td>Sales and Marketing</td>\n",
       "      <td>Analytics</td>\n",
       "      <td>[SUNDAY, SATURDAY]</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>53183116-2cb2-32ee-9042-d62eb7061407</td>\n",
       "      <td>2022-05-01</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.146806</td>\n",
       "      <td>0.520167</td>\n",
       "      <td>1.75</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>7.50</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Support</td>\n",
       "      <td>3</td>\n",
       "      <td>Individual Contributor</td>\n",
       "      <td>Sales and Marketing</td>\n",
       "      <td>IT</td>\n",
       "      <td>[SUNDAY, SATURDAY]</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 155 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0                              PersonId  MetricDate  \\\n",
       "0           1  a6afe34c-8524-32d3-a368-1517b29b68cd  2022-05-01   \n",
       "1           2  d6368140-9312-380b-bbc9-9a32bcef4b83  2022-05-01   \n",
       "2           3  60bf99b0-65fd-3c3f-94fb-8ceb451d59e7  2022-05-01   \n",
       "3           4  93fddd74-3667-392b-ba5a-92d855772cb0  2022-05-01   \n",
       "4           5  53183116-2cb2-32ee-9042-d62eb7061407  2022-05-01   \n",
       "\n",
       "   After_hours_call_hours  After_hours_chat_hours  \\\n",
       "0                     0.0                     0.0   \n",
       "1                     0.0                     0.0   \n",
       "2                     0.0                     0.0   \n",
       "3                     0.0                     0.0   \n",
       "4                     0.0                     0.0   \n",
       "\n",
       "   After_hours_collaboration_hours  After_hours_email_hours  \\\n",
       "0                        18.675938                 0.722722   \n",
       "1                         4.827803                 0.925556   \n",
       "2                         1.497806                 0.812806   \n",
       "3                        59.265892                 2.283668   \n",
       "4                         2.146806                 0.520167   \n",
       "\n",
       "   After_hours_meeting_hours  After_hours_scheduled_call_hours  \\\n",
       "0                      18.25                               0.0   \n",
       "1                       4.00                               0.0   \n",
       "2                       0.75                               0.0   \n",
       "3                      59.00                               0.0   \n",
       "4                       1.75                               0.0   \n",
       "\n",
       "   After_hours_unscheduled_call_hours  ...  Working_hours_meeting_hours  \\\n",
       "0                                   0  ...                        19.50   \n",
       "1                                   0  ...                         8.75   \n",
       "2                                   0  ...                        12.50   \n",
       "3                                   0  ...                        28.50   \n",
       "4                                   0  ...                         7.50   \n",
       "\n",
       "   Working_hours_scheduled_call_hours  Working_hours_unscheduled_call_hours  \\\n",
       "0                                   0                                     0   \n",
       "1                                   0                                     0   \n",
       "2                                   0                                     0   \n",
       "3                                   0                                     0   \n",
       "4                                   0                                     0   \n",
       "\n",
       "   LevelDesignation  Layer     SupervisorIndicator         Organization  \\\n",
       "0           Manager      3                 Manager  Sales and Marketing   \n",
       "1           Support      3  Individual Contributor              Finance   \n",
       "2           Support      3  Individual Contributor              Product   \n",
       "3          Director      2                Manager+  Sales and Marketing   \n",
       "4           Support      3  Individual Contributor  Sales and Marketing   \n",
       "\n",
       "   FunctionType         WeekendDays  IsActive  \n",
       "0       G_and_A  [SUNDAY, SATURDAY]      True  \n",
       "1         Sales  [SUNDAY, SATURDAY]      True  \n",
       "2            IT  [SUNDAY, SATURDAY]      True  \n",
       "3     Analytics  [SUNDAY, SATURDAY]      True  \n",
       "4            IT  [SUNDAY, SATURDAY]      True  \n",
       "\n",
       "[5 rows x 155 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pq_data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8939ef2c",
   "metadata": {},
   "source": [
    "## Calculating Information Value (IV)\n",
    "\n",
    "To run the IV methodology, a binary dependent variable is required. \n",
    "\n",
    "We can simulate such a variable by the following, and in this example we can name the variable `IsLargeNetwork`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "9e7ea31c",
   "metadata": {},
   "outputs": [],
   "source": [
    "pq_data[\"IsLargeNetwork\"] = np.where(pq_data[\"Internal_network_size\"] > 40, 1, 0)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "04d626a7",
   "metadata": {},
   "source": [
    "We can then define a list of predictors, and assign this to `predictor_list`. \n",
    "\n",
    "As shown below, `create_IV()` is the primary function for analyzing and visualizing Information Value for a selected outcome variable. We use the `predictors` argument to supply the list of predictors, and `outcome` to specify which varible to use as the dependent variable. \n",
    "\n",
    "In `return_type`, we specify a plot to be returned:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "59fefd00",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "predictor_list = [\n",
    "    \"Email_hours\",\n",
    "    \"Chat_hours\",\n",
    "    \"Meeting_hours\",\n",
    "    \"After_hours_collaboration_hours\",\n",
    "    \"Multitasking_hours\",\n",
    "    \"Meeting_and_call_hours_with_manager_1_1\"\n",
    "]\n",
    "\n",
    "\n",
    "vi.create_IV(\n",
    "    pq_data,\n",
    "    predictors = predictor_list,\n",
    "    outcome = \"IsLargeNetwork\",\n",
    "    return_type = \"plot\"\n",
    "    )"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8864fa41",
   "metadata": {},
   "source": [
    "Here's a general guideline on how to interpret the IV values:\n",
    "\n",
    "- IV < 0.02: The predictor is not useful for modeling (it has no predictive power).\n",
    "- 0.02 <= IV < 0.1: The predictor has only a weak predictive power.\n",
    "- 0.1 <= IV < 0.3: The predictor has a medium predictive power.\n",
    "- 0.3 <= IV < 0.5: The predictor has a strong predictive power.\n",
    "- IV >= 0.5: The predictor has a suspiciously high predictive power, and may potentially indicate overfitting. \n",
    "\n",
    "These are just guidelines and the thresholds can vary depending on the context and the specific problem you're working on. With real data, always consider the business context and use your judgement when interpreting the IV values. \n",
    "\n",
    "## Other return options\n",
    "\n",
    "In total, there are five return options that can be supplied to `create_IV()`, via `return_type`: \n",
    "\n",
    "- \"plot\"\n",
    "- \"summary\"\n",
    "- \"list\"\n",
    "- \"plot-WOE\"\n",
    "- \"IV\"\n",
    "\n",
    "The below shows the results when `return_type = 'summary'`, which returns a DataFrame containing one row per predictor and its associated IV and p-value:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "86f854f1",
   "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>Variable</th>\n",
       "      <th>IV</th>\n",
       "      <th>pval</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Email_hours</td>\n",
       "      <td>1.774995</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Multitasking_hours</td>\n",
       "      <td>1.248193</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Meeting_hours</td>\n",
       "      <td>0.944530</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>After_hours_collaboration_hours</td>\n",
       "      <td>0.586598</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Chat_hours</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.003346</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                          Variable        IV      pval\n",
       "0                      Email_hours  1.774995  0.000000\n",
       "1               Multitasking_hours  1.248193  0.000000\n",
       "2                    Meeting_hours  0.944530  0.000000\n",
       "3  After_hours_collaboration_hours  0.586598  0.000000\n",
       "4                       Chat_hours  0.000000  0.003346"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vi.create_IV(\n",
    "    pq_data,\n",
    "    predictors = predictor_list,\n",
    "    outcome = \"IsLargeNetwork\",\n",
    "    return_type = \"summary\"\n",
    "    )"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e37b7343",
   "metadata": {},
   "source": [
    "It's also possible to return Weight of Evidence (WoE) as a plot too. The WoE for a given interval is calculated as the natural logarithm of the proportion of positive outcomes to the proportion of negative outcomes. In other words, it measures the evidence in favor of a particular outcome given the value of the independent variable.\n",
    "\n",
    "Here is the output for `return_type = 'plot-WOE'`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "9e17de5e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "[None, None, None, None, None]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vi.create_IV(\n",
    "    pq_data,\n",
    "    predictors = predictor_list,\n",
    "    outcome = \"IsLargeNetwork\",\n",
    "    return_type = \"plot-WOE\"\n",
    "    )"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fc77305d",
   "metadata": {},
   "source": [
    "It's also possible to return more detailed outputs behind the calculations for `return_type = 'plot-WOE'`.\n",
    "\n",
    "When `return_type = 'IV'`, a list of three items is printed AND returned. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "e0b1c03b",
   "metadata": {},
   "outputs": [],
   "source": [
    "result_iv = vi.create_IV(\n",
    "    pq_data,\n",
    "    predictors = predictor_list,\n",
    "    outcome = \"IsLargeNetwork\",\n",
    "    return_type = \"IV\"\n",
    "    )"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "074fb00f",
   "metadata": {},
   "source": [
    "The first item in the list output is a dictionary of data frames that contain information about WOE, IV, odds, and probabilities. \n",
    "The second item in the list output is a DataFrame of IV and p-value, identical to the output in `return_type = 'summary'`.\n",
    "The third item in the list output is the natural log odds. \n",
    "\n",
    "You can extract them as follows: "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "45c5e5c8",
   "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>Email_hours</th>\n",
       "      <th>n</th>\n",
       "      <th>percentage</th>\n",
       "      <th>WOE</th>\n",
       "      <th>IV</th>\n",
       "      <th>ODDS</th>\n",
       "      <th>PROB</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[0.2,0.8]</td>\n",
       "      <td>200</td>\n",
       "      <td>0.2</td>\n",
       "      <td>-2.694161</td>\n",
       "      <td>0.675652</td>\n",
       "      <td>0.025641</td>\n",
       "      <td>0.025</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[0.8,1.0]</td>\n",
       "      <td>200</td>\n",
       "      <td>0.2</td>\n",
       "      <td>-1.071255</td>\n",
       "      <td>0.847590</td>\n",
       "      <td>0.129944</td>\n",
       "      <td>0.115</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[1.0,1.2]</td>\n",
       "      <td>200</td>\n",
       "      <td>0.2</td>\n",
       "      <td>-0.726511</td>\n",
       "      <td>0.935044</td>\n",
       "      <td>0.183432</td>\n",
       "      <td>0.155</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[1.2,1.5]</td>\n",
       "      <td>200</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.458575</td>\n",
       "      <td>0.981046</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>0.375</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[1.5,4.4]</td>\n",
       "      <td>200</td>\n",
       "      <td>0.2</td>\n",
       "      <td>1.840623</td>\n",
       "      <td>1.774995</td>\n",
       "      <td>2.389831</td>\n",
       "      <td>0.705</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Email_hours    n  percentage       WOE        IV      ODDS   PROB\n",
       "0   [0.2,0.8]  200         0.2 -2.694161  0.675652  0.025641  0.025\n",
       "1   [0.8,1.0]  200         0.2 -1.071255  0.847590  0.129944  0.115\n",
       "2   [1.0,1.2]  200         0.2 -0.726511  0.935044  0.183432  0.155\n",
       "3   [1.2,1.5]  200         0.2  0.458575  0.981046  0.600000  0.375\n",
       "4   [1.5,4.4]  200         0.2  1.840623  1.774995  2.389831  0.705"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Extract 'Email_hours' DataFrame from dictionary `result_iv[0]`:\n",
    "result_iv[0]['Email_hours']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "110ad8a7",
   "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>Variable</th>\n",
       "      <th>IV</th>\n",
       "      <th>pval</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Email_hours</td>\n",
       "      <td>1.774995</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Multitasking_hours</td>\n",
       "      <td>1.248193</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Meeting_hours</td>\n",
       "      <td>0.944530</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>After_hours_collaboration_hours</td>\n",
       "      <td>0.586598</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Chat_hours</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.003346</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                          Variable        IV      pval\n",
       "0                      Email_hours  1.774995  0.000000\n",
       "1               Multitasking_hours  1.248193  0.000000\n",
       "2                    Meeting_hours  0.944530  0.000000\n",
       "3  After_hours_collaboration_hours  0.586598  0.000000\n",
       "4                       Chat_hours  0.000000  0.003346"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result_iv[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "c4967af8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-0.9694005571881036"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result_iv[2]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a0a962a0",
   "metadata": {},
   "source": [
    "Here is a guide on interpreting WoE, odds, and probabilities: \n",
    "- A positive WoE value indicates that the odds of the event are higher for the group in question than for the entire dataset. In other words, the event is more likely to occur for this group.\n",
    "- A negative WoE value indicates that the odds of the event are lower for the group in question than for the entire dataset. In other words, the event is less likely to occur for this group.\n",
    "- A WoE of zero indicates that the odds of the event for the group are the same as for the entire dataset.\n",
    "\n",
    "**Odds**: The odds of an event occurring is the ratio of the probability of the event occurring to the probability of the event not occurring. \n",
    "\n",
    "**Probability**: This is the likelihood of the event occurring, a value between 0 and 1. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "62e34397",
   "metadata": {},
   "source": [
    "To return only this dictionary of DataFrames, you can also run `return_type = 'list'`, which returns the identical dictionary: "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "7afac66d",
   "metadata": {},
   "outputs": [],
   "source": [
    "result_iv_full = vi.create_IV(\n",
    "    pq_data,\n",
    "    predictors = predictor_list,\n",
    "    outcome = \"IsLargeNetwork\",\n",
    "    return_type = \"list\"\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "ddf716ad",
   "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>Email_hours</th>\n",
       "      <th>n</th>\n",
       "      <th>percentage</th>\n",
       "      <th>WOE</th>\n",
       "      <th>IV</th>\n",
       "      <th>ODDS</th>\n",
       "      <th>PROB</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[0.2,0.8]</td>\n",
       "      <td>200</td>\n",
       "      <td>0.2</td>\n",
       "      <td>-2.694161</td>\n",
       "      <td>0.675652</td>\n",
       "      <td>0.025641</td>\n",
       "      <td>0.025</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[0.8,1.0]</td>\n",
       "      <td>200</td>\n",
       "      <td>0.2</td>\n",
       "      <td>-1.071255</td>\n",
       "      <td>0.847590</td>\n",
       "      <td>0.129944</td>\n",
       "      <td>0.115</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[1.0,1.2]</td>\n",
       "      <td>200</td>\n",
       "      <td>0.2</td>\n",
       "      <td>-0.726511</td>\n",
       "      <td>0.935044</td>\n",
       "      <td>0.183432</td>\n",
       "      <td>0.155</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[1.2,1.5]</td>\n",
       "      <td>200</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.458575</td>\n",
       "      <td>0.981046</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>0.375</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[1.5,4.4]</td>\n",
       "      <td>200</td>\n",
       "      <td>0.2</td>\n",
       "      <td>1.840623</td>\n",
       "      <td>1.774995</td>\n",
       "      <td>2.389831</td>\n",
       "      <td>0.705</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Email_hours    n  percentage       WOE        IV      ODDS   PROB\n",
       "0   [0.2,0.8]  200         0.2 -2.694161  0.675652  0.025641  0.025\n",
       "1   [0.8,1.0]  200         0.2 -1.071255  0.847590  0.129944  0.115\n",
       "2   [1.0,1.2]  200         0.2 -0.726511  0.935044  0.183432  0.155\n",
       "3   [1.2,1.5]  200         0.2  0.458575  0.981046  0.600000  0.375\n",
       "4   [1.5,4.4]  200         0.2  1.840623  1.774995  2.389831  0.705"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result_iv_full['Email_hours']"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3f7ed03f",
   "metadata": {},
   "source": [
    "## Notes \n",
    "\n",
    "### Additional arguments\n",
    "\n",
    "There are two other arguments `create_IV()`, i.e. `siglevel` and `exc_sig` which controls whether significance results are shown in the outputs. These are optional. \n",
    "\n",
    "### Methodology choice\n",
    "\n",
    "When contemplating whether to use the Information Value methodology, it's worth noting that WoE has several advantages:\n",
    "\n",
    "1. It can transform a continuous variable into a set of categories, which can capture non-linear effects.\n",
    "1. It creates monotonic variables, which are often better handled by some statistical models.\n",
    "1. It allows you to compare the predictive power of variables from different scales and distributions.\n",
    "\n",
    "### Function architecture\n",
    "\n",
    "The `create_IV()` function calls a few other functions: \n",
    "\n",
    "  - `calculate_IV()`\n",
    "  - `map_IV()`\n",
    "  - `create_bar_asis()`\n",
    "  - `p_test()`\n"
   ]
  }
 ],
 "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": "3.12.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}