SYSTEMIQ Assessment.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "cf29e067",
   "metadata": {},
   "source": [
    "# SYSTEMIQ Assessment\n",
    "The following notebook contains my analysis for the assessment for the 'Python Modeller - Energy Platform' at SYSTEMIQ. Everything contained within this notebook is my own work.\n",
    "\n",
    "Victor Onink 26/04/2022"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "18f2e133",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from scipy.stats import ttest_ind\n",
    "from datetime import timedelta, datetime\n",
    "import statsmodels.api as sm"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "81d58e3e",
   "metadata": {},
   "source": [
    "### Data cleaning & merging\n",
    "1. Please load the users and activity tables into Python\n",
    "\n",
    "    a. How many variables are in the datasets\n",
    "    \n",
    "    b. How many observations are in the datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "6900d830",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The user dataset has 5807 observations and 5 variables.\n",
      "The activity dataset has 22642 observations and 3 variables.\n"
     ]
    }
   ],
   "source": [
    "user_data = pd.read_csv('users.csv')\n",
    "activity_data = pd.read_csv('activities.csv')\n",
    "\n",
    "def data_parameters(data_set, name) -> print:\n",
    "    \"\"\"\n",
    "    Print the number of observations and variables contained within the data_set. The data_set data is \n",
    "    structured according to [num_observations, num_variables]\n",
    "    \"\"\"\n",
    "    print('The {} dataset has {} observations and {} variables.'.format(name, *data_set.shape))\n",
    "    \n",
    "data_parameters(user_data, 'user')\n",
    "data_parameters(activity_data, 'activity')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "193dd327",
   "metadata": {},
   "source": [
    "2. How many:\n",
    "    \n",
    "    a. Male users are in the dataset?\n",
    "    \n",
    "    b. Female users are in the dataset?\n",
    "    \n",
    "    c. For haw many users is no gender information available?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "c9f7d5a0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The dataset has 2909 male users.\n",
      "The dataset has 1417 female users.\n",
      "The dataset has 1481 users without available gender information.\n"
     ]
    }
   ],
   "source": [
    "male_users = sum(user_data.Gender == 'M')\n",
    "female_users = sum(user_data.Gender == 'F')\n",
    "undefined_users = sum(user_data.Gender == ' ')\n",
    "\n",
    "print('The dataset has {} male users.'.format(male_users))\n",
    "print('The dataset has {} female users.'.format(female_users))\n",
    "print('The dataset has {} users without available gender information.'.format(undefined_users))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b207d26c",
   "metadata": {},
   "source": [
    "3. Merge the two datasets and calculate the days since registration for each activity. What is the total mean and median revenue generated per user (ie across all activities)? How do you interpret the result?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "c2a72883",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The mean revenue is 512.69, the median revenue is 80.00\n"
     ]
    }
   ],
   "source": [
    "# Merging the data, where the join is done on the 'UserId' variable\n",
    "merged_data = user_data.merge(activity_data, left_on='UserId', right_on='UserId')\n",
    "\n",
    "# Calculating the total revenue for each user\n",
    "total_revenue = merged_data.groupby('UserId').sum('Revenue')\n",
    "\n",
    "# Calculate the mean and median revenue per user\n",
    "mean_revenue = total_revenue.mean()\n",
    "median_revenue = total_revenue.median()\n",
    "\n",
    "print('The mean revenue is {:.2f}, the median revenue is {:.2f}'.format(mean_revenue.Revenue, median_revenue.Revenue))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "96acd955",
   "metadata": {},
   "source": [
    "As can be seen above, the mean revenue is significantly higher than the median. This suggests that there are a number of users in the dataset that spend significantly more money over the various activities in the dataset than others."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "aaa5e1af",
   "metadata": {},
   "source": [
    "4. Please visualize the distribution of the revenues by user."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "c90365b4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# We can just make a simple 1D histogram of the data, with revenueon x axis and number of users on the y axis\n",
    "def plot_histogram(revenue_data, mean_rev, median_rev) -> plt.figure:\n",
    "    \"\"\"\n",
    "    This function makes a standard histogram of the revenue_data, as well as indicating the mean and median \n",
    "    revenues (mean_rev, median_rev).\n",
    "    \"\"\"\n",
    "    fig = plt.figure(figsize=(15, 10))\n",
    "    ax = fig.add_subplot(111)\n",
    "    counts, bins, _ = ax.hist(revenue_data, bins=np.logspace(1, 5, num=100))\n",
    "\n",
    "    # Setting the axis labels\n",
    "    plt.xlabel('Revenue', fontsize=18)\n",
    "    plt.ylabel('Number of users', fontsize=18)\n",
    "    _ = plt.xticks(fontsize=15)\n",
    "    _ = plt.yticks(fontsize=15)\n",
    "\n",
    "    # Setting the axis limits and making the x axis a log scale\n",
    "    plt.xlim(10, 1e5)\n",
    "    plt.ylim(0, 600)\n",
    "    plt.xscale('log')\n",
    "\n",
    "    # Plotting the mean and median values\n",
    "    plt.axvline(x=mean_rev, color='tab:red', linewidth=5, label='Mean revenue')\n",
    "    plt.axvline(x=median_rev, color='black', linewidth=5, label='Median revenue')\n",
    "\n",
    "    _ = ax.legend(fontsize=15)\n",
    "    \n",
    "plot_histogram(total_revenue, mean_revenue.Revenue, median_revenue.Revenue)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fe82acfc",
   "metadata": {},
   "source": [
    "The figure above shows that the revenue per user distribution is highly skewed, which was already expected based on the large difference between the mean and median values. Half of the users spend less than 80 euro on activities, but there are a small number of users that spend 10,000 - 100,000 euro. As such, the mean revenue per user is 512.69 euro, even though the majority of the users spend signigicantly less than that."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4e60d5c6",
   "metadata": {},
   "source": [
    "5. What is the average week 1 revenue (ie the revenue generated by the user in the first week)?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "776ca027",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The week 1 mean revenue is 200.25, the median revenue is 50.00\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def get_period_events(all_data, time_period=7) -> pd.DataFrame:\n",
    "    \"\"\"\n",
    "    Returns a boolean DataFrame that indicates which events occur within time_period days of registration\n",
    "    \"\"\"\n",
    "    # Assure that the time columns are all in datetime format\n",
    "    all_data.RegDate = pd.to_datetime(merged_data.RegDate)\n",
    "    all_data.Date = pd.to_datetime(merged_data.Date)\n",
    "    \n",
    "    # Calculate the difference in days between the Date and the RegDate\n",
    "    days_difference = all_data.Date - all_data.RegDate \n",
    "    \n",
    "    # Calculate the revenue per user within time_period days of registration\n",
    "    return days_difference.dt.days <= time_period\n",
    "    \n",
    "\n",
    "week1_events_bool = get_period_events(merged_data)\n",
    "week1_events = merged_data.loc[week1_events_bool]\n",
    "week1_revenue = week1_events.groupby('UserId').sum('Revenue')\n",
    "\n",
    "# Calculate the mean and median revenue per user\n",
    "mean_week1_revenue = week1_revenue.mean()\n",
    "median_week1_revenue = week1_revenue.median()\n",
    "\n",
    "print('The week 1 mean revenue is {:.2f}, the median revenue is {:.2f}'.format(mean_week1_revenue.Revenue, \n",
    "                                                                               median_week1_revenue.Revenue))\n",
    "plot_histogram(week1_revenue, mean_week1_revenue.Revenue, median_week1_revenue.Revenue)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "44da0831",
   "metadata": {},
   "source": [
    "As we can see with the histogram above, if we just consider the mean/median revenue per user in their first week since registration, the averages are lower. There are still a number of high values (>10,000 euro), but they are less common. As such, the averages over the entire dataset are influenced by that a number of users have been using the service more longer periods of time, which is why the average revenue is higher. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4eb86c7d",
   "metadata": {},
   "source": [
    "### Analytics\n",
    "6. In this dataset men generated a higher week 1 revenue than women, on average.  Is this difference in revenue between men and women statistically significant? What is an appropriate statistical test to determine this and what is it’s p-value?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "2c042851",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The mean male revenue is 207.99±788.82 euro\n",
      "The mean female revenue is 174.69±576.33 euro\n",
      "the independent t-test value is 1.349, with significance 0.177\n"
     ]
    }
   ],
   "source": [
    "# First, we replace all missing gender values with 'M', as required by the assignment \n",
    "merged_data.Gender = np.where((merged_data.Gender == ' '), 'M', merged_data.Gender)\n",
    "\n",
    "# Next, split the data according to male/female\n",
    "male_data = merged_data[merged_data.Gender == 'M']\n",
    "female_data = merged_data[merged_data.Gender == 'F']\n",
    "\n",
    "# Now, we sort out all the events that are in the first week of registration\n",
    "male_days_difference = male_data.Date - male_data.RegDate\n",
    "female_days_difference = female_data.Date - female_data.RegDate\n",
    "male_week1_events = male_data[male_days_difference.dt.days <= 7]\n",
    "female_week1_events = female_data[female_days_difference.dt.days <= 7]\n",
    "\n",
    "# Calculate the revenue per person \n",
    "male_revenue = male_week1_events.groupby('UserId').sum('Revenue')\n",
    "female_revenue = female_week1_events.groupby('UserId').sum('Revenue')\n",
    "\n",
    "# Now, we can conduct the independent sample t test to compare the dataset\n",
    "t_value, p_value = ttest_ind(male_revenue.Revenue, female_revenue.Revenue)\n",
    "\n",
    "# Printing the results\n",
    "print('The mean male revenue is {:.2f}±{:.2f} euro'.format(male_revenue.Revenue.mean(), male_revenue.Revenue.std()))\n",
    "print('The mean female revenue is {:.2f}±{:.2f} euro'.format(female_revenue.Revenue.mean(), female_revenue.Revenue.std()))\n",
    "print('the independent t-test value is {:.3f}, with significance {:.3f}'.format(t_value, p_value))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5ac61a1d",
   "metadata": {},
   "source": [
    "The mean revenue for men and women is quite similar, especially given the large standard deviation on both of these values (as a result of the extremely high spenders shown in the histograms earlier). Given that male/female makes for two completely independent datasets, the independent sample t test is a suitable metric to see whether this difference is significant. With a significance value $p=0.177$, we can conclude that there is not a significant difference in the amount of revenue generated by male and female users."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "036d7b81",
   "metadata": {},
   "source": [
    "7. In which country is this difference greatest?\n",
    "8. Please visualise the relationship between country, gender and week 1 revenue with an appropriate chart."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "2a20fe60",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The absolute difference in revenue for men and women is each country is:\n",
      "           Revenue\n",
      "Country           \n",
      "DE       30.651792\n",
      "FR       28.066291\n",
      "GB       17.140430\n",
      "Other    70.622211\n",
      "US       63.705906\n",
      "So, we can conclude the largest difference is in Other, where men generate 159.05 and women 88.42\n"
     ]
    }
   ],
   "source": [
    "# We are going to take the male/female revenue and split them up into different countries\n",
    "male_country = male_week1_events.groupby(['Country', 'UserId']).sum('Revenue')\n",
    "female_country = female_week1_events.groupby(['Country', 'UserId']).sum('Revenue')\n",
    "\n",
    "# Next we calculate the average for each country\n",
    "male_country_mean = male_country.groupby('Country').mean('Revenue')\n",
    "female_country_mean = female_country.groupby('Country').mean('Revenue')\n",
    "\n",
    "# We also calculate the standard deviation\n",
    "male_country_std = male_country.groupby('Country').std()\n",
    "female_country_std = female_country.groupby('Country').std()\n",
    "\n",
    "\n",
    "# Calculate the difference\n",
    "difference_revenue = np.abs(male_country_mean - female_country_mean)\n",
    "\n",
    "print('The absolute difference in revenue for men and women is each country is:')\n",
    "print(difference_revenue)\n",
    "max_region = difference_revenue.idxmax().values[0]\n",
    "str_format = max_region, male_country_mean.Revenue.loc[max_region], female_country_mean.Revenue.loc[max_region]\n",
    "print('So, we can conclude the largest difference is in {}, where men generate {:.2f} and women {:.2f}'.format(*str_format))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "a9e7bef4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# To compare the values in each region, we will create a bar graph. For this, we first combine the two dataframes\n",
    "join_pd = pd.DataFrame({'Country': male_country_mean.index,\n",
    "                        'Male': male_country_mean.Revenue,\n",
    "                        'Female': female_country_mean.Revenue,\n",
    "                        'Male_std': male_country_std.Revenue,\n",
    "                        'Female_std': female_country_std.Revenue})\n",
    "\n",
    "# Create the figure\n",
    "plt.rc('axes', titlesize=16)     # fontsize of the axes title\n",
    "plt.rc('axes', labelsize=16)    # fontsize of the x and y labels\n",
    "plt.rc('xtick', labelsize=14)    # fontsize of the tick labels\n",
    "plt.rc('ytick', labelsize=14)    # fontsize of the tick labels\n",
    "plt.rc('legend', fontsize=16)    # legend fontsize\n",
    "\n",
    "fig = plt.figure(figsize=(15, 10))\n",
    "ax = fig.add_subplot(111)\n",
    "join_pd.plot(x=\"Country\", y=[\"Male\", \"Female\"], kind=\"bar\", ax=ax, yerr=join_pd[['Male_std', 'Female_std']].T.values)\n",
    "ax.set_ylim([0, 1300])\n",
    "_ = ax.set_ylabel('Mean revenue')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "19442b22",
   "metadata": {},
   "source": [
    "So, from the analysis we can see that men generally generate more mean revenue than women, and this is a trend that is visible over all regions included within the data set. The difference is largest in the 'Other' region, where the large difference here might be due to having a range of different regions being put together into one category. We can also see that all regions show high variability in the data, which is illustrated by the large uncertainty bars in the data."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c1082d08",
   "metadata": {},
   "source": [
    "9.\tBuild a linear regression model to predict the week 1 revenue of a user based on the following variables: Gender, Age, Country and day 1 revenue (ie the revenue generated by the user on their registration date). Which of these variables have a statistically significant predictive power according to this model?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "e3e7c192",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "       UserId        DOB     RegDate  Revenue  week1_revenue  day1_revenue  \\\n",
      "0     1326865 1988-05-04  2016-03-24   351.42           0.00          0.00   \n",
      "1     1333278 1971-07-21  2016-02-09    45.00           0.00          0.00   \n",
      "2     1336967 1983-07-31  2016-01-29    20.00           0.00          0.00   \n",
      "3     1339784 1976-09-21  2016-01-21    59.28           0.00          0.00   \n",
      "4     1339829 1981-05-03  2016-04-12    17.05           0.00          0.00   \n",
      "...       ...        ...         ...      ...            ...           ...   \n",
      "5792  3347222 1967-05-25  2017-03-22   230.27          56.80         37.52   \n",
      "5793  3347251 1981-06-11  2017-04-03    30.00          30.00         30.00   \n",
      "5794  3347333 1987-11-22  2017-01-27    30.00          30.00         30.00   \n",
      "5795  3347367 1995-08-09  2017-04-04    17.05          17.05         17.05   \n",
      "5796  3347450 1994-07-30  2017-04-21    42.63          42.63         42.63   \n",
      "\n",
      "       age  Gender_M  Country_FR  Country_GB  Country_Other  Country_US  \n",
      "0     33.0         1           0           0              0           0  \n",
      "1     50.0         1           0           1              0           0  \n",
      "2     38.0         1           0           1              0           0  \n",
      "3     45.0         1           0           0              1           0  \n",
      "4     40.0         0           0           0              1           0  \n",
      "...    ...       ...         ...         ...            ...         ...  \n",
      "5792  54.0         1           1           0              0           0  \n",
      "5793  40.0         0           0           1              0           0  \n",
      "5794  34.0         1           0           1              0           0  \n",
      "5795  26.0         0           0           0              0           1  \n",
      "5796  27.0         1           0           0              0           1  \n",
      "\n",
      "[5797 rows x 12 columns]\n"
     ]
    }
   ],
   "source": [
    "# First, we need to calculate the day 1 and week 1revenue, which can just follow the procedure we used for the \n",
    "# 7 day revenue\n",
    "day1_events_bool = get_period_events(merged_data, time_period=0)\n",
    "week1_events_bool = get_period_events(merged_data, time_period=7)\n",
    "\n",
    "# Now, we set the revenue for all events outside of 1 day or 1 week to 0\n",
    "day1_revenue, week1_revenue = merged_data.Revenue.copy(deep=True), merged_data.Revenue.copy(deep=True)\n",
    "day1_revenue[day1_events_bool == False] = 0\n",
    "week1_revenue[week1_events_bool == False] = 0\n",
    "\n",
    "# And now the mean revenue per person for week1 and day1\n",
    "merged_data['week1_revenue'] = week1_revenue\n",
    "merged_data['day1_revenue'] = day1_revenue\n",
    "user_mean_revenue = merged_data.groupby('UserId').sum(['week1_revenue', 'day1_revenue'])\n",
    "\n",
    "# Now we merge these revenues with the users\n",
    "complete_data = user_data.merge(user_mean_revenue, on='UserId')\n",
    "    \n",
    "# Finally, we calculate the age of the person relative to today\n",
    "today = datetime.now()\n",
    "complete_data.DOB = pd.to_datetime(complete_data.DOB)\n",
    "complete_data['age'] = (today - complete_data.DOB).dt.days // 365.25\n",
    "\n",
    "# And then we again make sure all unmarked genders are 'M', and then create dummy variables since the linear \n",
    "# regression model doesn't like string\n",
    "complete_data.Gender = np.where((complete_data.Gender == ' '), 'M', complete_data.Gender)\n",
    "complete_data = pd.get_dummies(complete_data, columns=['Gender'], drop_first=True)\n",
    "\n",
    "# Similarly, we generate dummy variables for the countries\n",
    "complete_data = pd.get_dummies(complete_data, columns=['Country'], drop_first=True)\n",
    "print(complete_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "30ef940c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>      <td>week1_revenue</td>  <th>  R-squared (uncentered):</th>      <td>   0.485</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared (uncentered):</th> <td>   0.484</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th>          <td>   777.8</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 26 Apr 2022</td> <th>  Prob (F-statistic):</th>           <td>  0.00</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>17:11:24</td>     <th>  Log-Likelihood:    </th>          <td> -44475.</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>  5797</td>      <th>  AIC:               </th>          <td>8.896e+04</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>  5790</td>      <th>  BIC:               </th>          <td>8.901e+04</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     7</td>      <th>                     </th>              <td> </td>    \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>              <td> </td>    \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "        <td></td>           <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Gender_M</th>      <td>    0.5074</td> <td>   14.131</td> <td>    0.036</td> <td> 0.971</td> <td>  -27.194</td> <td>   28.209</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>age</th>           <td>    0.6533</td> <td>    0.370</td> <td>    1.766</td> <td> 0.078</td> <td>   -0.072</td> <td>    1.379</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>day1_revenue</th>  <td>    1.4593</td> <td>    0.021</td> <td>   68.688</td> <td> 0.000</td> <td>    1.418</td> <td>    1.501</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Country_FR</th>    <td>   29.8458</td> <td>   28.415</td> <td>    1.050</td> <td> 0.294</td> <td>  -25.858</td> <td>   85.550</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Country_GB</th>    <td>   22.2315</td> <td>   18.005</td> <td>    1.235</td> <td> 0.217</td> <td>  -13.065</td> <td>   57.528</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Country_Other</th> <td>   -1.2415</td> <td>   22.490</td> <td>   -0.055</td> <td> 0.956</td> <td>  -45.330</td> <td>   42.847</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Country_US</th>    <td>   -2.6947</td> <td>   21.406</td> <td>   -0.126</td> <td> 0.900</td> <td>  -44.659</td> <td>   39.270</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>14676.084</td> <th>  Durbin-Watson:     </th>   <td>   1.970</td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th>  <td> 0.000</td>   <th>  Jarque-Bera (JB):  </th> <td>394547536.172</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>           <td>27.660</td>   <th>  Prob(JB):          </th>   <td>    0.00</td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td>1279.869</td>  <th>  Cond. No.          </th>   <td>1.71e+03</td>   \n",
       "</tr>\n",
       "</table><br/><br/>Notes:<br/>[1] R² is computed without centering (uncentered) since the model does not contain a constant.<br/>[2] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[3] The condition number is large, 1.71e+03. This might indicate that there are<br/>strong multicollinearity or other numerical problems."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                                 OLS Regression Results                                \n",
       "=======================================================================================\n",
       "Dep. Variable:          week1_revenue   R-squared (uncentered):                   0.485\n",
       "Model:                            OLS   Adj. R-squared (uncentered):              0.484\n",
       "Method:                 Least Squares   F-statistic:                              777.8\n",
       "Date:                Tue, 26 Apr 2022   Prob (F-statistic):                        0.00\n",
       "Time:                        17:11:24   Log-Likelihood:                         -44475.\n",
       "No. Observations:                5797   AIC:                                  8.896e+04\n",
       "Df Residuals:                    5790   BIC:                                  8.901e+04\n",
       "Df Model:                           7                                                  \n",
       "Covariance Type:            nonrobust                                                  \n",
       "=================================================================================\n",
       "                    coef    std err          t      P>|t|      [0.025      0.975]\n",
       "---------------------------------------------------------------------------------\n",
       "Gender_M          0.5074     14.131      0.036      0.971     -27.194      28.209\n",
       "age               0.6533      0.370      1.766      0.078      -0.072       1.379\n",
       "day1_revenue      1.4593      0.021     68.688      0.000       1.418       1.501\n",
       "Country_FR       29.8458     28.415      1.050      0.294     -25.858      85.550\n",
       "Country_GB       22.2315     18.005      1.235      0.217     -13.065      57.528\n",
       "Country_Other    -1.2415     22.490     -0.055      0.956     -45.330      42.847\n",
       "Country_US       -2.6947     21.406     -0.126      0.900     -44.659      39.270\n",
       "==============================================================================\n",
       "Omnibus:                    14676.084   Durbin-Watson:                   1.970\n",
       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):        394547536.172\n",
       "Skew:                          27.660   Prob(JB):                         0.00\n",
       "Kurtosis:                    1279.869   Cond. No.                     1.71e+03\n",
       "==============================================================================\n",
       "\n",
       "Notes:\n",
       "[1] R² is computed without centering (uncentered) since the model does not contain a constant.\n",
       "[2] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[3] The condition number is large, 1.71e+03. This might indicate that there are\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Now that we have all our variables of interest, we can create our linear regression model, with X being our\n",
    "# training data and y the target data\n",
    "X = complete_data[['Gender_M', 'age', 'day1_revenue', 'Country_FR', 'Country_GB', 'Country_Other', 'Country_US']]\n",
    "y = complete_data['week1_revenue']\n",
    "\n",
    "linear_model = sm.OLS(y, X).fit()\n",
    "linear_model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4b6e817e",
   "metadata": {},
   "source": [
    "So, if we look at the results above, we can see that only the `day1_revenue` has significant predictive power ($p<0.05$), although `age` is very close to significant. In contrast, the user gender has very little predictive power, which is in line with the earlier conclusion that the mean revenue generated by men and women is not significantly different. The nationality of the user is also not a signficant predictor, and this is also to be expected, given that the bar plot shows that all the male/female mean revenues fall within each other's standard deviation.\n",
    "\n",
    "In short, a model just using the user `age` and `day1_revenue` is likely the best predictive model that you could get from this dataset."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a1fc6fbb",
   "metadata": {},
   "source": [
    "10.\tWhat revenues do you expect women from France, Germany and the UK to generate, assuming they are all aged 40 and all generated £20 on their registration day?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "2e26f902",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0    85.164513\n",
      "1    77.550216\n",
      "2    55.318754\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "# To make these predictions, we first create a dataframe with the necessary data\n",
    "users = pd.DataFrame({'Gender_M': [0, 0, 0], 'age': [40, 40, 40], 'day1_revenue': [20, 20, 20],\n",
    "                      'Country_FR': [1, 0, 0], 'Country_GB': [0, 1, 0], 'Country_Other': [0, 0, 0],\n",
    "                      'Country_US': [0, 0, 0]})\n",
    "\n",
    "print(linear_model.predict(users))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "95a59d5e",
   "metadata": {},
   "source": [
    "So, based on this linear regression model, we would expect a 40 year old women in France, Germany and the UK respectively to generate 85.16, 55.32 and 77.55 pounds of revenue."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.6.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}