Simple Regression Model with Key Economic Indicators to predict Fed Funds Rate in United States.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "31c95956",
   "metadata": {},
   "source": [
    "# Simple Regression Model with Key Economic Indicators to predict Fed Funds Rate in United States"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "c2367b83",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn import metrics\n",
    "from sklearn.model_selection import train_test_split\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "96d9845b",
   "metadata": {},
   "source": [
    "This script was written in June 2022, using the latest dataset downloaded from FRED (Federal Reserve Economic Data website). The links are provided in the cell below.\n",
    "\n",
    "Backdrop:\n",
    "In relation to the present macroeconomic climate where inflation is at a 40 year high (8.6% from figures just released), due to both external (Ukraine war, China COVID lockdowns) and internal factors (robust US economy with generous monetary policy from Fed), it would be useful to examine how macro factors in the United States affect the Fed's decision to increase/decrease interest rates. It has been repeatedly mentioned by the Fed Chairman Jerome Powell in press conferences that the committee adopts a data-driven and empirical approach in determining the frequency and magnitude of interest rate changes. The recent hawkish shift in Fed policy is due to inflation running at a 40 year high which threatens to derail the US economy and has a profound impact on global economies. The Fed has repeatedly mentioned it is committed to fighting high inflation by raising interest rates from post pandemic lows to an appropriate level to taper demand while supply shortages are expected to remain elevated for some time.\n",
    "\n",
    "This program seeks to explore the relationship between the Effective Fed Funds Rate (target variable) and the most common parameters the Fed uses to justify their rate hikes/drop decisions when communicating to public, namely personal consumption expenditures inflation, real gross domestic production growth, and unemployment rate (predictor variables). We will predict the Theoretical Fed Funds Rates based on the predictor variables and compare that with the current interest rates, to have a general sensing of how far behind the Fed is lagging in their interest rate decision."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "ba352681",
   "metadata": {},
   "outputs": [],
   "source": [
    "# CSV files are taken directly from FRED, Short for Federal Reserve Economic Data, \n",
    "# FRED is an online database consisting of hundreds of thousands of economic data time series from scores of national, international, public, and private sources\n",
    "# https://fred.stlouisfed.org/graph/?g=LOaq\n",
    "# https://fred.stlouisfed.org/graph/?g=LOaz\n",
    "# https://fred.stlouisfed.org/graph/?g=LObH\n",
    "# https://fred.stlouisfed.org/graph/?g=LOaR"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "daf99f6f",
   "metadata": {},
   "source": [
    "## Preprocessing data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "65bc485d",
   "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>DATE</th>\n",
       "      <th>GDPC1_PC1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1948-01-01</td>\n",
       "      <td>2.60473</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1948-04-01</td>\n",
       "      <td>4.57732</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1948-07-01</td>\n",
       "      <td>5.39412</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1948-10-01</td>\n",
       "      <td>3.88678</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1949-01-01</td>\n",
       "      <td>0.93698</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>292</th>\n",
       "      <td>2021-01-01</td>\n",
       "      <td>0.54698</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>293</th>\n",
       "      <td>2021-04-01</td>\n",
       "      <td>12.22668</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>294</th>\n",
       "      <td>2021-07-01</td>\n",
       "      <td>4.94656</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>295</th>\n",
       "      <td>2021-10-01</td>\n",
       "      <td>5.53348</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>296</th>\n",
       "      <td>2022-01-01</td>\n",
       "      <td>3.54469</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>297 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           DATE  GDPC1_PC1\n",
       "0    1948-01-01    2.60473\n",
       "1    1948-04-01    4.57732\n",
       "2    1948-07-01    5.39412\n",
       "3    1948-10-01    3.88678\n",
       "4    1949-01-01    0.93698\n",
       "..          ...        ...\n",
       "292  2021-01-01    0.54698\n",
       "293  2021-04-01   12.22668\n",
       "294  2021-07-01    4.94656\n",
       "295  2021-10-01    5.53348\n",
       "296  2022-01-01    3.54469\n",
       "\n",
       "[297 rows x 2 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>DATE</th>\n",
       "      <th>UNRATE</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1948-01-01</td>\n",
       "      <td>3.7333333333333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1948-04-01</td>\n",
       "      <td>3.6666666666666667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1948-07-01</td>\n",
       "      <td>3.7666666666666667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1948-10-01</td>\n",
       "      <td>3.8333333333333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1949-01-01</td>\n",
       "      <td>4.6666666666666667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>293</th>\n",
       "      <td>2021-04-01</td>\n",
       "      <td>5.9000000000000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>294</th>\n",
       "      <td>2021-07-01</td>\n",
       "      <td>5.1000000000000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>295</th>\n",
       "      <td>2021-10-01</td>\n",
       "      <td>4.2333333333333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>296</th>\n",
       "      <td>2022-01-01</td>\n",
       "      <td>3.8000000000000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>297</th>\n",
       "      <td>2022-04-01</td>\n",
       "      <td>.</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>298 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           DATE              UNRATE\n",
       "0    1948-01-01  3.7333333333333333\n",
       "1    1948-04-01  3.6666666666666667\n",
       "2    1948-07-01  3.7666666666666667\n",
       "3    1948-10-01  3.8333333333333333\n",
       "4    1949-01-01  4.6666666666666667\n",
       "..          ...                 ...\n",
       "293  2021-04-01  5.9000000000000000\n",
       "294  2021-07-01  5.1000000000000000\n",
       "295  2021-10-01  4.2333333333333333\n",
       "296  2022-01-01  3.8000000000000000\n",
       "297  2022-04-01                   .\n",
       "\n",
       "[298 rows x 2 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>DATE</th>\n",
       "      <th>PCEPI_PC1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1960-01-01</td>\n",
       "      <td>1.69826</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1960-04-01</td>\n",
       "      <td>1.81333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1960-07-01</td>\n",
       "      <td>1.58048</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1960-10-01</td>\n",
       "      <td>1.47789</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1961-01-01</td>\n",
       "      <td>1.53312</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>245</th>\n",
       "      <td>2021-04-01</td>\n",
       "      <td>3.86105</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>246</th>\n",
       "      <td>2021-07-01</td>\n",
       "      <td>4.27777</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>247</th>\n",
       "      <td>2021-10-01</td>\n",
       "      <td>5.49727</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>248</th>\n",
       "      <td>2022-01-01</td>\n",
       "      <td>6.28677</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>249</th>\n",
       "      <td>2022-04-01</td>\n",
       "      <td>.</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>250 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           DATE PCEPI_PC1\n",
       "0    1960-01-01   1.69826\n",
       "1    1960-04-01   1.81333\n",
       "2    1960-07-01   1.58048\n",
       "3    1960-10-01   1.47789\n",
       "4    1961-01-01   1.53312\n",
       "..          ...       ...\n",
       "245  2021-04-01   3.86105\n",
       "246  2021-07-01   4.27777\n",
       "247  2021-10-01   5.49727\n",
       "248  2022-01-01   6.28677\n",
       "249  2022-04-01         .\n",
       "\n",
       "[250 rows x 2 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>DATE</th>\n",
       "      <th>DFF</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1954-07-01</td>\n",
       "      <td>1.44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1954-10-01</td>\n",
       "      <td>1.44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1955-01-01</td>\n",
       "      <td>1.44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1955-04-01</td>\n",
       "      <td>1.69</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1955-07-01</td>\n",
       "      <td>2.19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>267</th>\n",
       "      <td>2021-04-01</td>\n",
       "      <td>0.08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>268</th>\n",
       "      <td>2021-07-01</td>\n",
       "      <td>0.06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>269</th>\n",
       "      <td>2021-10-01</td>\n",
       "      <td>0.07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>270</th>\n",
       "      <td>2022-01-01</td>\n",
       "      <td>0.33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>271</th>\n",
       "      <td>2022-04-01</td>\n",
       "      <td>0.83</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>272 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           DATE   DFF\n",
       "0    1954-07-01  1.44\n",
       "1    1954-10-01  1.44\n",
       "2    1955-01-01  1.44\n",
       "3    1955-04-01  1.69\n",
       "4    1955-07-01  2.19\n",
       "..          ...   ...\n",
       "267  2021-04-01  0.08\n",
       "268  2021-07-01  0.06\n",
       "269  2021-10-01  0.07\n",
       "270  2022-01-01  0.33\n",
       "271  2022-04-01  0.83\n",
       "\n",
       "[272 rows x 2 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Reading in files ...\n",
    "real_gdp = pd.read_csv('Realgdp.csv')\n",
    "unemp = pd.read_csv('Unemp_rate.csv')\n",
    "pce = pd.read_csv('PCE.csv')\n",
    "fedfr = pd.read_csv('Fedfundsrate.csv')\n",
    "\n",
    "display(real_gdp)\n",
    "display(unemp)\n",
    "display(pce)\n",
    "display(fedfr)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "163698bb",
   "metadata": {},
   "source": [
    "### Exploratory Data Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "aaa78d62",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 249 entries, 0 to 248\n",
      "Data columns (total 5 columns):\n",
      " #   Column             Non-Null Count  Dtype  \n",
      "---  ------             --------------  -----  \n",
      " 0   date               249 non-null    object \n",
      " 1   fedfundrate        249 non-null    float64\n",
      " 2   pce                249 non-null    float64\n",
      " 3   real_gdp           249 non-null    float64\n",
      " 4   unemployment_rate  249 non-null    float64\n",
      "dtypes: float64(4), object(1)\n",
      "memory usage: 11.7+ KB\n"
     ]
    },
    {
     "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>fedfundrate</th>\n",
       "      <th>pce</th>\n",
       "      <th>real_gdp</th>\n",
       "      <th>unemployment_rate</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>249.000000</td>\n",
       "      <td>249.000000</td>\n",
       "      <td>249.000000</td>\n",
       "      <td>249.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>4.977791</td>\n",
       "      <td>3.261336</td>\n",
       "      <td>2.993748</td>\n",
       "      <td>5.981392</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>4.034908</td>\n",
       "      <td>2.417878</td>\n",
       "      <td>2.459866</td>\n",
       "      <td>1.665501</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.040000</td>\n",
       "      <td>-1.196590</td>\n",
       "      <td>-9.083740</td>\n",
       "      <td>3.400000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>1.910000</td>\n",
       "      <td>1.587940</td>\n",
       "      <td>1.768460</td>\n",
       "      <td>4.833333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>4.580000</td>\n",
       "      <td>2.513520</td>\n",
       "      <td>3.005890</td>\n",
       "      <td>5.700000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>6.830000</td>\n",
       "      <td>4.197810</td>\n",
       "      <td>4.367200</td>\n",
       "      <td>7.033333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>22.000000</td>\n",
       "      <td>11.511190</td>\n",
       "      <td>12.226680</td>\n",
       "      <td>12.966667</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       fedfundrate         pce    real_gdp  unemployment_rate\n",
       "count   249.000000  249.000000  249.000000         249.000000\n",
       "mean      4.977791    3.261336    2.993748           5.981392\n",
       "std       4.034908    2.417878    2.459866           1.665501\n",
       "min       0.040000   -1.196590   -9.083740           3.400000\n",
       "25%       1.910000    1.587940    1.768460           4.833333\n",
       "50%       4.580000    2.513520    3.005890           5.700000\n",
       "75%       6.830000    4.197810    4.367200           7.033333\n",
       "max      22.000000   11.511190   12.226680          12.966667"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.info()\n",
    "data.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "4233d614",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "date                 False\n",
       "fedfundrate          False\n",
       "pce                  False\n",
       "real_gdp             False\n",
       "unemployment_rate    False\n",
       "dtype: bool"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.isnull().any()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a17298ab",
   "metadata": {},
   "source": [
    "From above, none of the cells have null values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "ffa30958",
   "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>date</th>\n",
       "      <th>fedfundrate</th>\n",
       "      <th>pce</th>\n",
       "      <th>real_gdp</th>\n",
       "      <th>unemployment_rate</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1960-01-01</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1.69826</td>\n",
       "      <td>4.92542</td>\n",
       "      <td>5.133333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1960-04-01</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1.81333</td>\n",
       "      <td>2.05821</td>\n",
       "      <td>5.233333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1960-07-01</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.58048</td>\n",
       "      <td>2.48539</td>\n",
       "      <td>5.533333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1960-10-01</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.47789</td>\n",
       "      <td>0.88268</td>\n",
       "      <td>6.266667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1961-01-01</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.53312</td>\n",
       "      <td>-0.66913</td>\n",
       "      <td>6.800000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         date  fedfundrate      pce  real_gdp  unemployment_rate\n",
       "0  1960-01-01          4.0  1.69826   4.92542           5.133333\n",
       "1  1960-04-01          3.5  1.81333   2.05821           5.233333\n",
       "2  1960-07-01          3.0  1.58048   2.48539           5.533333\n",
       "3  1960-10-01          3.0  1.47789   0.88268           6.266667\n",
       "4  1961-01-01          2.0  1.53312  -0.66913           6.800000"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# merging different csvs into 1 dataframe\n",
    "data = real_gdp.merge(unemp, how='left', left_on='DATE', right_on='DATE')\n",
    "data = pce.merge(data, how='left', left_on='DATE', right_on='DATE')\n",
    "data = fedfr.merge(data, how='right', left_on='DATE', right_on='DATE')\n",
    "\n",
    "# renaming columns, where \n",
    "# pce = personal consumption expenditures inflation,\n",
    "# real_gdp = real gross domestic production growth,\n",
    "# unemployment_rate = unemployment rate\n",
    "# and fedfundrate = effective federal funds rate\n",
    "\n",
    "data.rename(columns={'DATE':'date', 'PCEPI_PC1': 'pce', 'DFF': 'fedfundrate', 'GDPC1_PC1': 'real_gdp', 'UNRATE': 'unemployment_rate'}, inplace=True)\n",
    "data.drop(labels=249, inplace=True)    # dropping last row due to invalid values in real_gdp/unemployment_rate inherent from source data\n",
    "\n",
    "data['pce'] = data['pce'].astype(float)                             # convert to float for further data processing\n",
    "data['unemployment_rate'] = data['unemployment_rate'].astype(float) # convert to float for further data processing\n",
    "\n",
    "data.head()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "e7d769ef",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.PairGrid at 0x17b628a1d30>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x360 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Some exploratory data visualization\n",
    "sns.pairplot(data=data, x_vars = [\"unemployment_rate\", \"pce\", \"real_gdp\"], y_vars = \"fedfundrate\", height=5, aspect=0.8, kind='reg')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "225b53f3",
   "metadata": {},
   "source": [
    "From above graphs, there appears to be little correlation between unemployment rate/real gdp versus Fed Funds Rate based on the spread of the data points versus the best fit line generated. \n",
    "\n",
    "PCE appears to be most correlated to Fed Funds Rate - when PCE increases, so does Fed Funds Rate. \n",
    "\n",
    "We will include all 3 predictors into the linear model to see how it affects the output."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a953b190",
   "metadata": {},
   "source": [
    "## Simple Linear Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "a58d9054",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LinearRegression()\n",
      "intercept:  -0.3140261983822201\n",
      "coefficients : [0.10679938 1.17323769 0.27613768]\n"
     ]
    }
   ],
   "source": [
    "X = data.loc[:, [\"unemployment_rate\", \"pce\", \"real_gdp\"]]                  \n",
    "y = data.loc[:, \"fedfundrate\"]\n",
    "\n",
    "# instantiate Linear Regression model\n",
    "lm = LinearRegression()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "2e6794a6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     unemployment_rate      pce  real_gdp\n",
      "180           5.300000  2.62657   3.90862\n",
      "70            6.900000  6.76472   5.77090\n",
      "190           4.666667  2.17711   2.39856\n",
      "     unemployment_rate       pce  real_gdp\n",
      "56            5.133333   9.05166   0.63877\n",
      "127           7.100000   2.49994   1.16650\n",
      "60            8.266667  10.33008  -2.29905\n",
      "LinearRegression()\n",
      "intercept:  -0.20563210540030852\n",
      "coefficients : [0.08483166 1.29009159 0.1738911 ]\n",
      "\n",
      "Train MSE: 7.584733494879259\n",
      "Train RMSE: 2.754039486804657\n",
      "Train R2: 0.5357685937261293\n",
      "\n",
      "Test MSE: 10.296472115526932\n",
      "Test RMSE: 3.2088116360308425\n",
      "Test R2: 0.340819578626705\n"
     ]
    }
   ],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state = 10)  \n",
    "print(X_train.head(3))\n",
    "print(X_test.head(3))\n",
    "\n",
    "lm.fit(X_train, y_train)                                  # to train the model\n",
    "print(lm)\n",
    "\n",
    "print(\"intercept: \",lm.intercept_)                         # print the coefficients\n",
    "print(\"coefficients :\", lm.coef_)\n",
    "print('')\n",
    "\n",
    "#Calculate MSE, RMSE, R2 scores for train set\n",
    "y_pred_train = lm.predict(X_train)\n",
    "print('Train MSE:', metrics.mean_squared_error(y_true=y_train, y_pred=y_pred_train))        #MSE, y_true are the real data points, y_pred is y_predict which is the regression line\n",
    "print('Train RMSE:', np.sqrt(metrics.mean_squared_error(y_true=y_train, y_pred=y_pred_train)))   #RMSE\n",
    "print('Train R2:', metrics.r2_score(y_train, y_pred_train))\n",
    "print('')\n",
    "\n",
    "#Calculate MSE, RMSE, R2 scores for test set\n",
    "y_pred_test = lm.predict(X_test)\n",
    "print('Test MSE:', metrics.mean_squared_error(y_true=y_test, y_pred=y_pred_test))        #MSE, y_true are the real data points, y_pred is y_predict which is the regression line\n",
    "print('Test RMSE:', np.sqrt(metrics.mean_squared_error(y_true=y_test, y_pred=y_pred_test)))   #RMSE\n",
    "print('Test R2:', metrics.r2_score(y_test, y_pred_test))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8433d7f4",
   "metadata": {},
   "source": [
    "R2 train and test scores are around 0.34 to 0.53 due to the large spread of values. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9e039c82",
   "metadata": {},
   "source": [
    "### Question: Based on the fitted model, using today's data, what would be the theoretical Fed funds rate?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "46ae562b",
   "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>date</th>\n",
       "      <th>fedfundrate</th>\n",
       "      <th>pce</th>\n",
       "      <th>real_gdp</th>\n",
       "      <th>unemployment_rate</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>244</th>\n",
       "      <td>2021-01-01</td>\n",
       "      <td>0.06</td>\n",
       "      <td>1.82982</td>\n",
       "      <td>0.54698</td>\n",
       "      <td>6.200000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>245</th>\n",
       "      <td>2021-04-01</td>\n",
       "      <td>0.08</td>\n",
       "      <td>3.86105</td>\n",
       "      <td>12.22668</td>\n",
       "      <td>5.900000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>246</th>\n",
       "      <td>2021-07-01</td>\n",
       "      <td>0.06</td>\n",
       "      <td>4.27777</td>\n",
       "      <td>4.94656</td>\n",
       "      <td>5.100000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>247</th>\n",
       "      <td>2021-10-01</td>\n",
       "      <td>0.07</td>\n",
       "      <td>5.49727</td>\n",
       "      <td>5.53348</td>\n",
       "      <td>4.233333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>248</th>\n",
       "      <td>2022-01-01</td>\n",
       "      <td>0.33</td>\n",
       "      <td>6.28677</td>\n",
       "      <td>3.54469</td>\n",
       "      <td>3.800000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           date  fedfundrate      pce  real_gdp  unemployment_rate\n",
       "244  2021-01-01         0.06  1.82982   0.54698           6.200000\n",
       "245  2021-04-01         0.08  3.86105  12.22668           5.900000\n",
       "246  2021-07-01         0.06  4.27777   4.94656           5.100000\n",
       "247  2021-10-01         0.07  5.49727   5.53348           4.233333\n",
       "248  2022-01-01         0.33  6.28677   3.54469           3.800000"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.tail()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "52ed43e2",
   "metadata": {},
   "source": [
    "We extract the predictor variables (unemployment rate, pce and real_gdp) for Jan 2022 and predict the Fed Funds Rate based on the linear model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "32d5a6bb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Theoretical Fed Funds Rate: 8.843627336774643\n"
     ]
    }
   ],
   "source": [
    "# the figures below correspond to the \"unemployment_rate\", \"pce\", \"real_gdp\" of Jan 2022\n",
    "y_pred = lm.predict(np.array([3.8, 6.28677, 3.54469]).reshape(1,-1))\n",
    "print('Theoretical Fed Funds Rate:', y_pred[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f8f68b3a",
   "metadata": {},
   "source": [
    "Based on our very simple model, the theoretical fed funds rate based on January 2022's macroeconomic figures is 8.8%! However, the Fed has only as of June 2022 raised interest rates to 1.5 to 1.75% and have much catching up to do. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "25637ea6",
   "metadata": {},
   "source": [
    "## Re-Run model using PCE as predictor"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "918f4081",
   "metadata": {},
   "source": [
    "We will re-run the model using only PCE as the predictor just to see how the resutls vary."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "3a29972e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LinearRegression()\n",
      "intercept:  1.194594759873978\n",
      "coefficients : [1.16001447]\n"
     ]
    }
   ],
   "source": [
    "X = data.loc[:, \"pce\"]                  \n",
    "y = data.loc[:, \"fedfundrate\"]\n",
    "# display(X)\n",
    "\n",
    "X = np.array(X).reshape(-1,1)\n",
    "\n",
    "# instantiate and fit\n",
    "lm = LinearRegression()\n",
    "lm.fit(X, y)                                  # to train the model\n",
    "print(lm)\n",
    "\n",
    "# print the coefficients\n",
    "print(\"intercept: \",lm.intercept_)\n",
    "print(\"coefficients :\", lm.coef_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "fd9158e5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Theoretical Fed Funds Rate: 8.48733889925574\n"
     ]
    }
   ],
   "source": [
    "x = np.array([6.28677]).reshape(1,-1)\n",
    "y_pred = lm.predict(x)\n",
    "print('Theoretical Fed Funds Rate:', y_pred[0])\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "696be999",
   "metadata": {},
   "source": [
    "The theoretical Fed Funds Rate is now 8.5%, not too far from the earlier predicted value of 8.8%. This is expected because PCE shows the highest correlation to Fed Funds Rate as compared to other predictors. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "55d6cb8f",
   "metadata": {},
   "source": [
    "### Conclusion"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "80f4453b",
   "metadata": {},
   "source": [
    "In conclusion, it is clear from the above data that there is a significant disparity in current interest rates (1.5 to 1.75%) as compared to the theoretical values (~8%). Coincidentally, the theoretical interest rate also matches roughly the current inflation rate, which makes some sense given that the Fed had once raised interest rates to as high as 20% to fight inflation (in the double dights) in the 1980s. The Fed has also repeatedly telegraphed that they would be more aggressive in their monetary policy to combat inflation, which does not seem to be slowing down at the moment. The Fed now projects that the median interest rates would rise to about 3.8% in 2023. While still a far cry from our theoretical value, this may be due to the Fed adopting a wait and see approach for the economy to cool substantially after the planned series of rate hikes in 2022/2023 and easing of supply chain related factors in time to come. Nonetheless, while it can be hard to predict exactly where interest rates end up, I expect the trajectory of the interest rates to be on the uptrend and it would be prudent to take this into account when making any business / investment related decisions."
   ]
  }
 ],
 "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": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}