Adsdata_modeling_v3.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "dab92afa-a8d3-4a06-a763-dc7e2741d671",
   "metadata": {},
   "source": [
    "**How Should Companies Adjust their Advertisement Budget to Increase Product Sales?**"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7af7a414-22e2-49d4-9e4d-71c08320ccdd",
   "metadata": {},
   "source": [
    "In this project, I will demonstrate how *could a company adjust their advertisement budget* to increase or maximize their sales of a particular product using an example dataset (see the source below). This data consists of the sales of that product in 200 different markets, along with advertising budgets for TV, radio, and newspaper for the product for each of the markets. **Note that here we assume the company doesn't know how to increase sales of the product.** Our task is to potentially find association between ads budget and sales. In order to accomplish this, we could develop a model that can be used to predict sales using the three different media budget.\n",
    "\n",
    "We first download Numpy and Pandas packages to Python. Then, we the import the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "9ebbb99f-fdf2-48b1-98b7-d4e7f3bfddbb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "        TV  radio  newspaper  sales\n",
      "0    230.1   37.8       69.2   22.1\n",
      "1     44.5   39.3       45.1   10.4\n",
      "2     17.2   45.9       69.3    9.3\n",
      "3    151.5   41.3       58.5   18.5\n",
      "4    180.8   10.8       58.4   12.9\n",
      "..     ...    ...        ...    ...\n",
      "195   38.2    3.7       13.8    7.6\n",
      "196   94.2    4.9        8.1    9.7\n",
      "197  177.0    9.3        6.4   12.8\n",
      "198  283.6   42.0       66.2   25.5\n",
      "199  232.1    8.6        8.7   13.4\n",
      "\n",
      "[200 rows x 4 columns]\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "# We import the data from a csv file\n",
    "data = pd.read_csv(\"/Users/100355063/Desktop/Python_data/adsdata.csv\")\n",
    "\n",
    "# We drop the empthy column that comes with the data using the argument. You can manipulate the following to exlude any columns you don't need.\n",
    "df = pd.DataFrame(data.drop(['Unnamed: 0'], axis = 1))\n",
    "\n",
    "# We then print our data\n",
    "print(df)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "94f58ecd-de3d-4967-b10c-f846ba5c6e9a",
   "metadata": {},
   "source": [
    "--> You notice that in our data, sales is in thousands of units while TV, radio, and newspaper are in thousands of dollars."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2ea8a73b-d059-4375-a757-bc12b6669009",
   "metadata": {},
   "source": [
    "---\n",
    "Next, we plot each of the predictor variables (ads budgets) vs sales to get a little bit of insights into the data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "ca3f4b5f-4da3-4ec1-82b5-10f4bf0d5c45",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# data\n",
    "data = pd.read_csv(\"/Users/100355063/Desktop/Python_data/adsdata.csv\")\n",
    "df = pd.DataFrame(data.drop(['Unnamed: 0'], axis = 1))\n",
    "\n",
    "# Plot scatter plot\n",
    "plt.scatter(df['radio'], df['sales'])\n",
    "\n",
    "# Add axis labels and title\n",
    "plt.xlabel('Radio Ads Budget')\n",
    "plt.ylabel('Product Sales')\n",
    "plt.title('Scatter Plot of Radio vs Sales')\n",
    "\n",
    "# Show plot\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "f453c44a-b071-42b0-9c61-fa730e97330f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# data\n",
    "data = pd.read_csv(\"/Users/100355063/Desktop/Python_data/adsdata.csv\")\n",
    "df = pd.DataFrame(data.drop(['Unnamed: 0'], axis = 1))\n",
    "\n",
    "# Plot scatter plot\n",
    "plt.scatter(df['TV'], df['sales'])\n",
    "\n",
    "# Add axis labels and title\n",
    "plt.xlabel('TV Ads Budget')\n",
    "plt.ylabel('Product Sales')\n",
    "plt.title('Scatter Plot of TV vs Sales')\n",
    "\n",
    "# Show plot\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "e5684beb-a67a-4aa9-9815-ed009dff013a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# data\n",
    "data = pd.read_csv(\"/Users/100355063/Desktop/Python_data/adsdata.csv\")\n",
    "df = pd.DataFrame(data.drop(['Unnamed: 0'], axis = 1))\n",
    "\n",
    "# Plot scatter plot\n",
    "plt.scatter(df['newspaper'], df['sales'])\n",
    "\n",
    "# Add axis labels and title\n",
    "plt.xlabel('Newspaper Ads Budget')\n",
    "plt.ylabel('Product Sales')\n",
    "plt.title('Scatter Plot of Newspaper vs Sales')\n",
    "\n",
    "# Show plot\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "964cd1c7-92ce-4630-9987-9f47806dcc53",
   "metadata": {},
   "source": [
    "Based on the plots, it is somewhat clear to see a linear (or straight-ish) trend with TV and Radio vs sales but hard to observe that with newspaper vs sales (last plot). We definitely need to run some in-depth statistical anaylses to be able to analyze the data better and make confident conclusions.\n",
    "\n",
    "Next, we list the following questions to guide this project to complete our task."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "115d4504-b56f-4f8c-ac8f-69ccb28c9aa4",
   "metadata": {},
   "source": [
    "---\n",
    "**Question #1: Is there a relationship between product sales and advertising budget?** \n",
    "\n",
    "We could answer this question by fitting a multiple linear regression model using our advertising data. We regress sales (our response variable) onto TV, radio, and newspaper budgets (predictor variables). We install the following tools onto Python to be able to create a model:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "24e06483-a3f8-4432-a3f7-59f94226230d",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:                  sales   R-squared:                       0.897\n",
      "Model:                            OLS   Adj. R-squared:                  0.896\n",
      "Method:                 Least Squares   F-statistic:                     570.3\n",
      "Date:                Tue, 14 Mar 2023   Prob (F-statistic):           1.58e-96\n",
      "Time:                        14:04:43   Log-Likelihood:                -386.18\n",
      "No. Observations:                 200   AIC:                             780.4\n",
      "Df Residuals:                     196   BIC:                             793.6\n",
      "Df Model:                           3                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "Intercept      2.9389      0.312      9.422      0.000       2.324       3.554\n",
      "TV             0.0458      0.001     32.809      0.000       0.043       0.049\n",
      "radio          0.1885      0.009     21.893      0.000       0.172       0.206\n",
      "newspaper     -0.0010      0.006     -0.177      0.860      -0.013       0.011\n",
      "==============================================================================\n",
      "Omnibus:                       60.414   Durbin-Watson:                   2.084\n",
      "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              151.241\n",
      "Skew:                          -1.327   Prob(JB):                     1.44e-33\n",
      "Kurtosis:                       6.332   Cond. No.                         454.\n",
      "==============================================================================\n",
      "\n",
      "Notes:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import LinearRegression\n",
    "import statsmodels.formula.api as smf\n",
    "\n",
    "model = smf.ols('sales ~ TV + radio + newspaper', data=df)\n",
    "model = model.fit()\n",
    "\n",
    "print(model.summary())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "15e3947f-6e82-4594-8005-8a1fdb8fa04b",
   "metadata": {},
   "source": [
    "--> According to our table, F-statistic is around **570** which is pretty high, so this suggests that at least one of our variables is associated with sales. In addition, R-squared of this model ended up being approximately **0.90** and so we suspect that variability in sales **could be explained by this multiple linear model.**\n",
    "\n",
    "Next, we must define our variables using letters for simplicity of the future analysis. We convert our data frame columns into Numphy arrays:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "1a11dfeb-25ca-43a8-a3da-13b123ee94c3",
   "metadata": {},
   "outputs": [],
   "source": [
    "y = np.array(df['sales'])\n",
    "x1 = np.array(df['TV'])\n",
    "x2 = np.array(df['radio'])\n",
    "x3 = np.array(df['newspaper'])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a02a3aa1-a3ec-4c54-b8c4-7386935173f5",
   "metadata": {},
   "source": [
    "--> Then we compute mean values for our response value (y), sales. The average value is 14.022 which is 14,022 units for the 200 markets provided in the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "69c3511c-0c21-424d-b740-6b8014a0ae49",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "14.0225"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from numpy import mean\n",
    "\n",
    "mean(y)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d84ad3eb-eb4c-43d1-9bf4-08180430b73f",
   "metadata": {},
   "source": [
    "---\n",
    "**Question #2: How strong is the relationship between sales and ads budget?**"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3b13cadf-2710-4f34-8049-8bb4130adf9d",
   "metadata": {},
   "source": [
    "In order to answer this question, we could look at two simple statistical indicators- RSE (residual standard error) estimates and R^2 statistic. \n",
    "\n",
    "We first compute RSE using the following commands on Python:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "99d8dab0-bfdb-4508-8c94-b6d1e0070c96",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.6727572743844112"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.resid.std()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cc49bdbe-b724-4dcb-bf57-17ad1588a288",
   "metadata": {},
   "source": [
    "--> We can say that on average, actual sales in each market deviate from the true regression line by approximately 1,680 units. Considering the average is 14,022 our percentage error is approximately %12 which is not too bad (note that this was computed by 1,680/14,022).\n",
    "\n",
    "Next we consider R-squared of this model, which ended up being approximately 0.90 (this is drawn from the uupper right portion of the regression model summary from above). Since this value is close to 1, we can confidently say that **90% of the variability in sales** could be explained by this linear regression model. This is a strong association!"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "86423270-2cbe-4fb9-be70-454f522866c4",
   "metadata": {},
   "source": [
    "---\n",
    "**Question #3: Which media adverting budget are associated with sales of the product?**\n",
    "\n",
    "In order to answer this question, we need to take a look at the regression table (provided above) and examine p-values and coefficient values attached to each of the predictor variables. Since the p-values for TV and Radio are very small, and for newspaper is high, we can say that **only TV and radio** budgets are associated with the sales for the product. \n",
    "\n",
    "We now take a look at the regression table again or just use Python's \"model.params\" function to pull coefficients for each of the predictors. The coefficient value for radio budget is 0.19 (rounded) meaning that every 1000 dollars increase in radio budget leads to an increase in sales by approximately **190 units** when TV and newspaper budget is fixed. Same thing could be said for TV budget- every 1000 dollars increase in that budget leads to an increase in sales by approximately **46 units** adjusted for the other two ads budgets."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "021a69a0-9c44-47c2-8376-a50a81e56fba",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Intercept    2.938889\n",
      "TV           0.045765\n",
      "radio        0.188530\n",
      "newspaper   -0.001037\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "print(model.params)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d59b169e-b7ff-4d20-8fb9-bd5588cbd000",
   "metadata": {},
   "source": [
    "---\n",
    "**Question #4: How large is the association between each media budget and sales?**"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fe367aa8-0614-41f0-8a9b-42ce12651b9e",
   "metadata": {},
   "source": [
    "In order to answer this question, we can take a look at confidence intervals (CI) of each of the predictor variables. Below we compute 95% CI for the coefficients in the model using Python:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "73ea3fbd-4d4b-4700-ad89-b1c54c6dd3a6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                  0         1\n",
      "Intercept  2.323762  3.554016\n",
      "TV         0.043014  0.048516\n",
      "radio      0.171547  0.205513\n",
      "newspaper -0.012616  0.010541\n"
     ]
    }
   ],
   "source": [
    "model = smf.ols('y ~ TV + radio + newspaper', data=df)\n",
    "model = model.fit()\n",
    "\n",
    "print(model.conf_int(0.05))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "56e611d5-e261-4e35-8736-ccece0605a8a",
   "metadata": {},
   "source": [
    "According to the print, we have CIs of (0.043, 0.049) for TV, (0.172, 0.206) for radio, and (−0.013, 0.011) for newspaper. This suggests that, \n",
    "- Using CI for radio, for each 1000 dollars increase in radio ads budget, the sales will see an average increase between 172 and 206 units. \n",
    "- Using CI for TV spending, for each 1000 dollars increase in TV spending, the sales will see an average increase between 43 and 48 units.\n",
    "- Since the interval for newspaper has a negative value, this variable is not statistically significant in the presence of the other two variables. \n",
    "- We could also interpret the CI for intercept as follows- sales will fall somewhere between 2,324 and 3,554 units in the absence of all the three media advertising.\n",
    "\n",
    "--> Another consideration when it comes to the association between the predictors (ads buudget) and response variable (sales) is the idea of correlation between predictors. Since we do not want predictors to be correlated with each other, we simply create a correlation matrix using the following function:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "id": "3bf5b6fb-7674-4f8e-b0bf-98d9c873b075",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                 TV     radio  newspaper     sales\n",
      "TV         1.000000  0.054809   0.056648  0.782224\n",
      "radio      0.054809  1.000000   0.354104  0.576223\n",
      "newspaper  0.056648  0.354104   1.000000  0.228299\n",
      "sales      0.782224  0.576223   0.228299  1.000000\n"
     ]
    }
   ],
   "source": [
    "corr_matrix = df.corr()\n",
    "print(corr_matrix)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8b682b51-552a-4025-8592-de41ffbf37fb",
   "metadata": {},
   "source": [
    "According to the table, newspaper and radio are positively correlated though it's a mild correlation. This indicates that **markets with high newspaper advertising tend to also have high radio advertising.**\n",
    "\n",
    "--> Now we are suspicious about potential **collinearity** between predictors which is situation in which two or more predictor variables are related to one another. To examine for this, we compute **variance inflation factor (VIF)**. The following is used to define and compute VIF:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "id": "f06b15ca-7a6b-4574-909e-21dd542a8231",
   "metadata": {},
   "outputs": [],
   "source": [
    "def calculate_vif(df, features):    \n",
    "    vif, tolerance = {}, {}\n",
    "    \n",
    "    # all the features that you want to examine\n",
    "    for feature in features:\n",
    "        # extract all the other features you will regress against\n",
    "        X = [f for f in features if f != feature]        \n",
    "        X, y = df[X], df[feature]\n",
    "        \n",
    "        # extract r-squared from the fit\n",
    "        r2 = LinearRegression().fit(X, y).score(X, y)                \n",
    "        \n",
    "        # calculate tolerance\n",
    "        tolerance[feature] = 1 - r2\n",
    "        \n",
    "        # calculate VIF\n",
    "        vif[feature] = 1/(tolerance[feature])\n",
    "        \n",
    "    # return VIF DataFrame\n",
    "    return pd.DataFrame({'VIF': vif, 'Tolerance': tolerance})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "id": "7f4ffa59-d3d1-42a0-bee8-3499f7773ef3",
   "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>VIF</th>\n",
       "      <th>Tolerance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>TV</th>\n",
       "      <td>1.004611</td>\n",
       "      <td>0.995410</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>radio</th>\n",
       "      <td>1.144952</td>\n",
       "      <td>0.873399</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>newspaper</th>\n",
       "      <td>1.145187</td>\n",
       "      <td>0.873220</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                VIF  Tolerance\n",
       "TV         1.004611   0.995410\n",
       "radio      1.144952   0.873399\n",
       "newspaper  1.145187   0.873220"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "calculate_vif(df=df, features=['TV','radio','newspaper'])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0a8443ae-3b6b-4387-a40a-79a8ca3abd31",
   "metadata": {},
   "source": [
    "Looking at the table above, we **do not** see anything problematic since all the VIF values are lower than 5. The mild correlation between radio and newspaper doesn't pose collinearity issue in the model.\n",
    "\n",
    "--> Next we can assess the association of each media budget individually on sales and so we would need to perform three separate simple linear regressions (instead of one big multiple linear regression including all our predictor variables). Here are the Python codes for that:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "id": "d3ff33a8-168d-4377-87bb-ddd330999aae",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model 1:\n",
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:                  sales   R-squared:                       0.612\n",
      "Model:                            OLS   Adj. R-squared:                  0.610\n",
      "Method:                 Least Squares   F-statistic:                     312.1\n",
      "Date:                Tue, 14 Mar 2023   Prob (F-statistic):           1.47e-42\n",
      "Time:                        14:04:43   Log-Likelihood:                -519.05\n",
      "No. Observations:                 200   AIC:                             1042.\n",
      "Df Residuals:                     198   BIC:                             1049.\n",
      "Df Model:                           1                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "Intercept      7.0326      0.458     15.360      0.000       6.130       7.935\n",
      "TV             0.0475      0.003     17.668      0.000       0.042       0.053\n",
      "==============================================================================\n",
      "Omnibus:                        0.531   Durbin-Watson:                   1.935\n",
      "Prob(Omnibus):                  0.767   Jarque-Bera (JB):                0.669\n",
      "Skew:                          -0.089   Prob(JB):                        0.716\n",
      "Kurtosis:                       2.779   Cond. No.                         338.\n",
      "==============================================================================\n",
      "\n",
      "Notes:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
      "\n",
      "Model 2:\n",
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:                  sales   R-squared:                       0.332\n",
      "Model:                            OLS   Adj. R-squared:                  0.329\n",
      "Method:                 Least Squares   F-statistic:                     98.42\n",
      "Date:                Tue, 14 Mar 2023   Prob (F-statistic):           4.35e-19\n",
      "Time:                        14:04:43   Log-Likelihood:                -573.34\n",
      "No. Observations:                 200   AIC:                             1151.\n",
      "Df Residuals:                     198   BIC:                             1157.\n",
      "Df Model:                           1                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "Intercept      9.3116      0.563     16.542      0.000       8.202      10.422\n",
      "radio          0.2025      0.020      9.921      0.000       0.162       0.243\n",
      "==============================================================================\n",
      "Omnibus:                       19.358   Durbin-Watson:                   1.946\n",
      "Prob(Omnibus):                  0.000   Jarque-Bera (JB):               21.910\n",
      "Skew:                          -0.764   Prob(JB):                     1.75e-05\n",
      "Kurtosis:                       3.544   Cond. No.                         51.4\n",
      "==============================================================================\n",
      "\n",
      "Notes:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
      "\n",
      "Model 3:\n",
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:                  sales   R-squared:                       0.052\n",
      "Model:                            OLS   Adj. R-squared:                  0.047\n",
      "Method:                 Least Squares   F-statistic:                     10.89\n",
      "Date:                Tue, 14 Mar 2023   Prob (F-statistic):            0.00115\n",
      "Time:                        14:04:43   Log-Likelihood:                -608.34\n",
      "No. Observations:                 200   AIC:                             1221.\n",
      "Df Residuals:                     198   BIC:                             1227.\n",
      "Df Model:                           1                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "Intercept     12.3514      0.621     19.876      0.000      11.126      13.577\n",
      "newspaper      0.0547      0.017      3.300      0.001       0.022       0.087\n",
      "==============================================================================\n",
      "Omnibus:                        6.231   Durbin-Watson:                   1.983\n",
      "Prob(Omnibus):                  0.044   Jarque-Bera (JB):                5.483\n",
      "Skew:                           0.330   Prob(JB):                       0.0645\n",
      "Kurtosis:                       2.527   Cond. No.                         64.7\n",
      "==============================================================================\n",
      "\n",
      "Notes:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import LinearRegression\n",
    "import statsmodels.formula.api as smf\n",
    "\n",
    "model1 = smf.ols('sales ~ TV', data=df)\n",
    "model1 = model1.fit()\n",
    "model2 = smf.ols('sales ~ radio', data=df)\n",
    "model2 = model2.fit()\n",
    "model3 = smf.ols('sales ~ newspaper', data=df)\n",
    "model3 = model3.fit()\n",
    "\n",
    "print(\"Model 1:\")\n",
    "print(model1.summary())\n",
    "print(\"\\nModel 2:\")\n",
    "print(model2.summary())\n",
    "print(\"\\nModel 3:\")\n",
    "print(model3.summary())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7ce8b71c-57d5-4a36-9f35-f6c27661e008",
   "metadata": {},
   "source": [
    "In Model 1, the association between TV and sales is strong, as indicated by the high coefficient estimate (0.0475) and the low p-value (0.000) for TV. The R-squared value (0.612) also suggests that the predictor TV explains a large portion of the variation in the response sales.\n",
    "\n",
    "In Model 2, the association between radio and sales is also strong, as indicated by the coefficient estimate (0.2025) and the low p-value (0.000) for radio. However, the R-squared value (0.332) suggests that the predictor radio explains a smaller portion of the variation in the response sales compared to TV in Model 1.\n",
    "\n",
    "In Model 3, the association between newspaper and sales is weaker, as indicated by the coefficient estimate (0.0547) and the higher p-value (0.001) for newspaper. The low R-squared value (0.052) suggests that the predictor newspaper explains a very small portion of the variation in the response sales, even when the values of TV and radio are ignored."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "853dba1f-e7e0-4f31-ae3b-b754a7e2505a",
   "metadata": {},
   "source": [
    "---\n",
    "**Question #5 - How accurately could we predict future sales using model?**\n",
    "\n",
    "We can compute response variable using our multiple linear regression. \n",
    "- If we want to compute individual response, we need to use a prediction intervals.\n",
    "- If we want to compute average response, we need to use confidence intervals.\n",
    "\n",
    "We are removing the predictor variable newspaper for predictions here since it has a very little impact. Let's assume 100,000 dollars is spent on TV advertising and 20,000 dollars is spent on radio advertising in each country. We need to calculate the **predicted sales** at these advertising levels (dollar inputs) and compute the confidence interval for the predicted sales. We can use the following steps:\n",
    "\n",
    "Identify the coefficients and intercept from the regression table: Intercept: 2.9211, TV: 0.0458, radio: 0.1880. Then we plug in the advertising levels into the regression equation to calculate the predicted sales:\n",
    "- predicted sales = intercept + TV * 100,000 + radio * 20,000\n",
    "- predicted sales = 2.9211 + 0.0458 * 100,000 + 0.1880 * 20,000\n",
    "- predicted sales = **11,786**\n",
    "\n",
    "Then we calculate the standard error of the predicted sales (s.e. = sqrt(MSE * (1/n + (TV - mean(TV))^2 / SS(TV) + (radio - mean(radio))^2 / SS(radio)))). We can use the mean and SS values from the regression table to compute the s.e.\n",
    "- mean(TV) = 147.0425 & SS(TV) = 1.6856e+07\n",
    "- mean(radio) = 23.264 & SS(radio) = 1846.92\n",
    "s.e. = sqrt(2.784 / 197 + (100,000 - 147.0425)^2 / 1.6856e+07 + (20,000 - 23.264)^2 / 1846.92)\n",
    "**Thus, s.e. = 897.079**\n",
    "\n",
    "Now we calculate the confidence interval using the t-distribution with 197 degrees of freedom and a 95% confidence level:\n",
    "\n",
    "CI = predicted sales +/- t(0.025, 197) * s.e.\n",
    "t(0.025, 197) = 1.972, CI = 11,786 +/- 1.972 * 897.079, **CI = (10,985, 11,587).**\n",
    "\n",
    "Therefore, the 95% confidence interval for the predicted sales at 100,000 dollars spent on TV advertising and 20,000 dollars spent on radio advertising is (10,985, 11,587). **This means that we can be 95% confident that the true value of the average sales in a large number of countries will fall within this interval.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "id": "fc6603d7-0816-42ab-93af-f5df0dcb404e",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "       Lower CI  Upper CI\n",
      "const  2.340343  3.501857\n",
      "TV     0.043013  0.048497\n",
      "radio  0.172139  0.203850\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import statsmodels.api as sm\n",
    "\n",
    "# Load the data\n",
    "data = pd.read_csv(\"/Users/100355063/Desktop/Python_data/adsdata.csv\")\n",
    "\n",
    "# Fit the OLS model\n",
    "X = data[['TV', 'radio']]\n",
    "X = sm.add_constant(X)\n",
    "y = data['sales']\n",
    "model = sm.OLS(y, X).fit()\n",
    "\n",
    "# Compute confidence intervals for the coefficients\n",
    "conf_int = model.conf_int(alpha=0.05)\n",
    "conf_int.columns = ['Lower CI', 'Upper CI']\n",
    "print(conf_int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "id": "ab3c3309-4acc-46e5-b67e-6328f2994f22",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:                  sales   R-squared:                       0.897\n",
      "Model:                            OLS   Adj. R-squared:                  0.896\n",
      "Method:                 Least Squares   F-statistic:                     859.6\n",
      "Date:                Tue, 14 Mar 2023   Prob (F-statistic):           4.83e-98\n",
      "Time:                        14:04:43   Log-Likelihood:                -386.20\n",
      "No. Observations:                 200   AIC:                             778.4\n",
      "Df Residuals:                     197   BIC:                             788.3\n",
      "Df Model:                           2                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "Intercept      2.9211      0.294      9.919      0.000       2.340       3.502\n",
      "TV             0.0458      0.001     32.909      0.000       0.043       0.048\n",
      "radio          0.1880      0.008     23.382      0.000       0.172       0.204\n",
      "==============================================================================\n",
      "Omnibus:                       60.022   Durbin-Watson:                   2.081\n",
      "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              148.679\n",
      "Skew:                          -1.323   Prob(JB):                     5.19e-33\n",
      "Kurtosis:                       6.292   Cond. No.                         425.\n",
      "==============================================================================\n",
      "\n",
      "Notes:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import LinearRegression\n",
    "import statsmodels.formula.api as smf\n",
    "\n",
    "model = smf.ols('sales ~ TV + radio', data=df)\n",
    "model = model.fit()\n",
    "\n",
    "print(model.summary())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d8dec960-2889-49bd-b7f4-46d7d2200647",
   "metadata": {},
   "source": [
    "--- \n",
    "**Question #6 - Is there an interaction effect?**\n",
    "\n",
    "It is always helpful to look at interaction effects of the predictor variables in addition to main effects. Previously, we regressed TV, Radio and Newspaper ads budgets onto sales. We noticed that newspaper didn't have a big impact so we removed that. But then we noticed that Radio and TV budgets were positively correlated so we suspect that those two variables might have some form on interaction (or synergy) effects. Here we create a new regression model:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "id": "acd22e7d-4573-4ce7-a6a6-4d6e779ed8c8",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:                  sales   R-squared:                       0.968\n",
      "Model:                            OLS   Adj. R-squared:                  0.967\n",
      "Method:                 Least Squares   F-statistic:                     1963.\n",
      "Date:                Tue, 14 Mar 2023   Prob (F-statistic):          6.68e-146\n",
      "Time:                        14:04:43   Log-Likelihood:                -270.14\n",
      "No. Observations:                 200   AIC:                             548.3\n",
      "Df Residuals:                     196   BIC:                             561.5\n",
      "Df Model:                           3                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "Intercept      6.7502      0.248     27.233      0.000       6.261       7.239\n",
      "TV             0.0191      0.002     12.699      0.000       0.016       0.022\n",
      "radio          0.0289      0.009      3.241      0.001       0.011       0.046\n",
      "TV:radio       0.0011   5.24e-05     20.727      0.000       0.001       0.001\n",
      "==============================================================================\n",
      "Omnibus:                      128.132   Durbin-Watson:                   2.224\n",
      "Prob(Omnibus):                  0.000   Jarque-Bera (JB):             1183.719\n",
      "Skew:                          -2.323   Prob(JB):                    9.09e-258\n",
      "Kurtosis:                      13.975   Cond. No.                     1.80e+04\n",
      "==============================================================================\n",
      "\n",
      "Notes:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
      "[2] The condition number is large, 1.8e+04. This might indicate that there are\n",
      "strong multicollinearity or other numerical problems.\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import LinearRegression\n",
    "import statsmodels.formula.api as smf\n",
    "\n",
    "model = smf.ols('sales ~ TV + radio + TV*radio', data=df)\n",
    "model = model.fit()\n",
    "\n",
    "print(model.summary())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "01cb1e7c-6c7c-4a86-b66a-996f571482ed",
   "metadata": {},
   "source": [
    "--> Now we have the model with including the interaction effect of radio and TV advertising. The p-value for the interaction term is low so that effect is significant. The overall R squared for this model is 0.96 vs 0.89 on the previous model (with all the predictor variables)) which means that the variation in sales is **explained better with the later model.**\n",
    "\n",
    "The interpretion of this is that every $1000 increase in TV adversiting is associated with an increase of 19 + 1.1 * radio advertising many units. Similar interpretation could be made for radio advertising. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "060d6570-15b7-4f78-b5f8-ffbc0a329f1d",
   "metadata": {},
   "source": [
    "Overall, we can learn a lot about this data through applying various statistical concepts and using regression models. It is clear that this sort of analysis couuld be helpful foor a company to make adjustment to help increase their product sales and increase profits. \n",
    "\n",
    "Gareth James, Daniela Witten, Trevor Hastie, and Robert Tibshirani's 2021 book titled \"An Introduction to Statistical Learning\" was mainly used for this project. This book is free and accessable through: https://www.statlearning.com/. The data for this project comes from: https://www.statlearning.com/resources-second-edition under Data Sets and Advertising.csv"
   ]
  }
 ],
 "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": "3.9.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}