Tesla_unit_root.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from scipy import stats\n",
    "\n",
    "import math\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "# Pandas options\n",
    "pd.set_option('display.max_colwidth', 1000, 'display.max_rows', 300, 'display.max_columns', None)\n",
    "\n",
    "# Plotting options\n",
    "%matplotlib inline\n",
    "\n",
    "import statsmodels.api as sm\n",
    "from statsmodels.tsa.stattools import adfuller\n",
    "from statsmodels.tsa.stattools import kpss\n",
    "\n",
    "from datetime import datetime, timedelta"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Part 1 - Load data and data processing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_data(path = 'TSLA.csv'):\n",
    "    \n",
    "    '''\n",
    "    \n",
    "    Load data\n",
    "\n",
    "    Args:path\n",
    "\n",
    "    Returns: dataframe   \n",
    "    \n",
    "    \n",
    "    '''\n",
    "    \n",
    "    \n",
    "    tsla = pd.read_csv(path, index_col=['Index'], parse_dates=['Index'])\n",
    "    tsla.index = tsla.index.tz_localize('Europe/Moscow')\n",
    "    tsla.index = tsla.index.tz_convert('America/New_York')\n",
    "    tsla = tsla.between_time('9:30', '16:00').reset_index()\n",
    "    tsla['day'] = pd.to_datetime(tsla.Index.dt.date)\n",
    "    tsla['time'] = tsla.Index.dt.strftime('%H:%M:%S')\n",
    "    tsla['qtr'] = tsla.Index.dt.to_period('Q')\n",
    "    tsla['month'] = tsla.Index.dt.to_period('M')\n",
    "    tsla.rename(columns = {'TSLA.Close': 'price'}, inplace = True)\n",
    "    cols = ['Index', 'day', 'time', 'qtr', 'month', 'price']\n",
    "    tsla = tsla[cols]\n",
    "    print(tsla.groupby('month')['price'].count().head(20))\n",
    "    print(tsla.shape)\n",
    "    tsla = tsla[tsla.day >= '2017-04-01'] # remove the month with a few observatiions\n",
    "    print(tsla.shape)\n",
    "    \n",
    "    # check the jump in each month and TSLA split up - 31 August 2020, move the observation on 2020.08.31 to the cohort of 2020.09 \n",
    "    x = pd.Series(np.datetime64('2020-09-01 00:00:00.000000000')).dt.to_period('M')\n",
    "    tsla.loc[(tsla.day >= '2020-08-31') & (tsla.day < '2020-10-01') , 'month'] = x[0]\n",
    "    \n",
    "    \n",
    "    return tsla"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "month\n",
      "2015-09       3\n",
      "2015-10       7\n",
      "2015-11      27\n",
      "2015-12      19\n",
      "2016-01      25\n",
      "2016-02      73\n",
      "2016-03      84\n",
      "2016-04     102\n",
      "2016-05     129\n",
      "2016-06     128\n",
      "2016-07     104\n",
      "2016-08     124\n",
      "2016-09     108\n",
      "2016-10      83\n",
      "2016-11     115\n",
      "2016-12     167\n",
      "2017-01     165\n",
      "2017-02     179\n",
      "2017-03     178\n",
      "2017-04    1370\n",
      "Freq: M, Name: price, dtype: int64\n",
      "(75137, 6)\n",
      "(73317, 6)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "tsla = load_data(path = 'TSLA.csv')\n",
    "# data visulization 1\n",
    "tsla['price'].plot(figsize=(15,5), title='TSLA Price', style='.')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Part 2 - Break the data and Formalising"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def break_data(data, period):\n",
    "    \n",
    "    '''  \n",
    "    divide the data by month\n",
    "\n",
    "    Args:\n",
    "    data - dataframe\n",
    "    period - month\n",
    "\n",
    "    Returns: dataframe     \n",
    "    \n",
    "    '''\n",
    "        \n",
    "        \n",
    "    df = data[data.month == period]\n",
    "    \n",
    "    # print(df.groupby('day').price.count().head())\n",
    "    \n",
    "    day = df.day.unique()\n",
    "    time =  df.time.unique()\n",
    "    D = len(day)\n",
    "    Obs = len(time)\n",
    "    DAY = np.repeat(day,Obs)\n",
    "    TIME = np.tile(time, D)\n",
    "    \n",
    "    tmp = pd.DataFrame({'day': DAY, 'time':TIME})\n",
    "    final = df.merge(tmp, how = 'outer')\n",
    "    final['Index'] = pd.to_datetime(final['day'].astype(str) + ' ' + final['time'].astype(str))\n",
    "    final['qtr'] = final.Index.dt.to_period('Q')\n",
    "    final['month'] = final.Index.dt.to_period('M')\n",
    "    final['price_log'] = np.log(final['price'])\n",
    "    final = final.sort_values(['Index'])\n",
    "    final['price_log_pre'] = final['price_log'].shift(1)\n",
    "    final['log_return'] = final['price_log'] - final['price_log_pre']\n",
    "    \n",
    "    \n",
    "    return final\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>Index</th>\n",
       "      <th>day</th>\n",
       "      <th>time</th>\n",
       "      <th>qtr</th>\n",
       "      <th>month</th>\n",
       "      <th>price</th>\n",
       "      <th>price_log</th>\n",
       "      <th>price_log_pre</th>\n",
       "      <th>log_return</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2018-04-02 09:30:00</td>\n",
       "      <td>2018-04-02</td>\n",
       "      <td>09:30:00</td>\n",
       "      <td>2018Q2</td>\n",
       "      <td>2018-04</td>\n",
       "      <td>254.60</td>\n",
       "      <td>5.539694</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2018-04-02 09:35:00</td>\n",
       "      <td>2018-04-02</td>\n",
       "      <td>09:35:00</td>\n",
       "      <td>2018Q2</td>\n",
       "      <td>2018-04</td>\n",
       "      <td>252.75</td>\n",
       "      <td>5.532401</td>\n",
       "      <td>5.539694</td>\n",
       "      <td>-0.007293</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2018-04-02 09:40:00</td>\n",
       "      <td>2018-04-02</td>\n",
       "      <td>09:40:00</td>\n",
       "      <td>2018Q2</td>\n",
       "      <td>2018-04</td>\n",
       "      <td>249.60</td>\n",
       "      <td>5.519860</td>\n",
       "      <td>5.532401</td>\n",
       "      <td>-0.012541</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2018-04-02 09:45:00</td>\n",
       "      <td>2018-04-02</td>\n",
       "      <td>09:45:00</td>\n",
       "      <td>2018Q2</td>\n",
       "      <td>2018-04</td>\n",
       "      <td>248.80</td>\n",
       "      <td>5.516649</td>\n",
       "      <td>5.519860</td>\n",
       "      <td>-0.003210</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2018-04-02 09:50:00</td>\n",
       "      <td>2018-04-02</td>\n",
       "      <td>09:50:00</td>\n",
       "      <td>2018Q2</td>\n",
       "      <td>2018-04</td>\n",
       "      <td>247.86</td>\n",
       "      <td>5.512864</td>\n",
       "      <td>5.516649</td>\n",
       "      <td>-0.003785</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                Index        day      time     qtr    month   price  \\\n",
       "0 2018-04-02 09:30:00 2018-04-02  09:30:00  2018Q2  2018-04  254.60   \n",
       "1 2018-04-02 09:35:00 2018-04-02  09:35:00  2018Q2  2018-04  252.75   \n",
       "2 2018-04-02 09:40:00 2018-04-02  09:40:00  2018Q2  2018-04  249.60   \n",
       "3 2018-04-02 09:45:00 2018-04-02  09:45:00  2018Q2  2018-04  248.80   \n",
       "4 2018-04-02 09:50:00 2018-04-02  09:50:00  2018Q2  2018-04  247.86   \n",
       "\n",
       "   price_log  price_log_pre  log_return  \n",
       "0   5.539694            NaN         NaN  \n",
       "1   5.532401       5.539694   -0.007293  \n",
       "2   5.519860       5.532401   -0.012541  \n",
       "3   5.516649       5.519860   -0.003210  \n",
       "4   5.512864       5.516649   -0.003785  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "final = break_data(tsla, period = '2018Q2')\n",
    "final.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Part 3 - DF statistics without jump"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def DF_Statistics(data):\n",
    "    \n",
    "    data = data.dropna()\n",
    "    n = data.shape[0]\n",
    "    \n",
    "    log_return = data['log_return']\n",
    "    log_price_pre = data['price_log_pre']\n",
    "    \n",
    "    alpha = np.ones(n)  # alpha = 1\n",
    "    beta = np.arange(n)  # beta*t = [0, 1, 2, ...]\n",
    "    \n",
    "    X = np.column_stack((log_price_pre, alpha))\n",
    "    results = sm.OLS(log_return, X).fit()\n",
    "    \n",
    "    return results.tvalues[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-2.5516555887320047"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "DF_Statistics(final) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Part 4 - Jump Detection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "def jump_detection(df, returns = 'log_return', alpha = 0.75):\n",
    "    \n",
    "    '''  \n",
    "    Detect jumps\n",
    "\n",
    "    Args:\n",
    "    df - dataframe\n",
    "    returns - \n",
    "    alpha\n",
    "\n",
    "    Returns: \n",
    "    res - dataframe\n",
    "    \n",
    "    '''\n",
    "        \n",
    "        \n",
    "    day = df.day.unique()\n",
    "    time =  df.time.unique()\n",
    "    D = len(day) # no. of days\n",
    "    Obs = len(time) # no. of observation in each day\n",
    "    \n",
    "    log_return = df[returns]\n",
    "    N = len(log_return)\n",
    "    \n",
    "    r = [np.nan] * N\n",
    "    m = [np.nan] * N\n",
    "    r_star = [np.nan] * N\n",
    "    r_star_mul = [np.nan] * N\n",
    "    BV = [np.nan] * N\n",
    "    sigma = [np.nan] * N\n",
    "    U = [np.nan] * N\n",
    "    M = int(np.sqrt(Obs)) # Look at past M value\n",
    "    i = math.ceil(M)\n",
    "    \n",
    "    while i <= N-1:\n",
    "        r[i] = log_return[i]\n",
    "        # print(r[i])\n",
    "        m[i] = np.nanmedian(log_return[(i-M):i])\n",
    "        # print(m[i])\n",
    "        r_star[i] = r[i] - m[i]\n",
    "        # print(r_star[i])\n",
    "        r_star_mul[i] = abs(r_star[i-1]) * abs(r_star[i]) if r_star[i-1] != np.nan else np.nan\n",
    "        # print(r_star_mul[i])\n",
    "        BV[i] = np.nansum(r_star_mul[(i-M+2):(i+1)]) * M / (M-1) * np.pi / 2 if r_star_mul[i] != np.nan else np.nan\n",
    "        # print(BV[i])\n",
    "        sigma[i] = np.sqrt(BV[i] / M) if BV[i] != np.nan else np.nan\n",
    "        # print(sigma[i])\n",
    "        U[i] = r_star[i] / sigma[i] if sigma[i] != np.nan else np.nan\n",
    "        # print(U[i])\n",
    "        i = i+ 1\n",
    "        # print(i)\n",
    "        # print('---------')\n",
    "    \n",
    "    U_abs = list(map(abs, U))\n",
    "    L = N\n",
    "    CV = (2 * np.log(L))**(1/2) - 1/2 * (2*np.log(L))**(-1/2)*(np.log(4*np.pi) + np.log(np.log(L))) + (2 * np.log(L))**(-1/2) * (-np.log(-np.log(1-alpha/2)))\n",
    "    jumps = (U_abs >= CV)\n",
    "    res = df[['day', 'time']]\n",
    "    res['jump'] = jumps\n",
    "    \n",
    "    res = df.merge(res, how = 'inner')\n",
    "    res['jump'] = res['jump'].astype(int)\n",
    "    res = res.dropna()\n",
    "    \n",
    "    \n",
    "    return res \n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "jumps = jump_detection(final, returns = 'log_return', alpha = 0.75)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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>Index</th>\n",
       "      <th>day</th>\n",
       "      <th>time</th>\n",
       "      <th>qtr</th>\n",
       "      <th>month</th>\n",
       "      <th>price</th>\n",
       "      <th>price_log</th>\n",
       "      <th>price_log_pre</th>\n",
       "      <th>log_return</th>\n",
       "      <th>jump</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2018-04-02 09:35:00</td>\n",
       "      <td>2018-04-02</td>\n",
       "      <td>09:35:00</td>\n",
       "      <td>2018Q2</td>\n",
       "      <td>2018-04</td>\n",
       "      <td>252.75</td>\n",
       "      <td>5.532401</td>\n",
       "      <td>5.539694</td>\n",
       "      <td>-0.007293</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2018-04-02 09:40:00</td>\n",
       "      <td>2018-04-02</td>\n",
       "      <td>09:40:00</td>\n",
       "      <td>2018Q2</td>\n",
       "      <td>2018-04</td>\n",
       "      <td>249.60</td>\n",
       "      <td>5.519860</td>\n",
       "      <td>5.532401</td>\n",
       "      <td>-0.012541</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2018-04-02 09:45:00</td>\n",
       "      <td>2018-04-02</td>\n",
       "      <td>09:45:00</td>\n",
       "      <td>2018Q2</td>\n",
       "      <td>2018-04</td>\n",
       "      <td>248.80</td>\n",
       "      <td>5.516649</td>\n",
       "      <td>5.519860</td>\n",
       "      <td>-0.003210</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2018-04-02 09:50:00</td>\n",
       "      <td>2018-04-02</td>\n",
       "      <td>09:50:00</td>\n",
       "      <td>2018Q2</td>\n",
       "      <td>2018-04</td>\n",
       "      <td>247.86</td>\n",
       "      <td>5.512864</td>\n",
       "      <td>5.516649</td>\n",
       "      <td>-0.003785</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2018-04-02 09:55:00</td>\n",
       "      <td>2018-04-02</td>\n",
       "      <td>09:55:00</td>\n",
       "      <td>2018Q2</td>\n",
       "      <td>2018-04</td>\n",
       "      <td>249.86</td>\n",
       "      <td>5.520901</td>\n",
       "      <td>5.512864</td>\n",
       "      <td>0.008037</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                Index        day      time     qtr    month   price  \\\n",
       "1 2018-04-02 09:35:00 2018-04-02  09:35:00  2018Q2  2018-04  252.75   \n",
       "2 2018-04-02 09:40:00 2018-04-02  09:40:00  2018Q2  2018-04  249.60   \n",
       "3 2018-04-02 09:45:00 2018-04-02  09:45:00  2018Q2  2018-04  248.80   \n",
       "4 2018-04-02 09:50:00 2018-04-02  09:50:00  2018Q2  2018-04  247.86   \n",
       "5 2018-04-02 09:55:00 2018-04-02  09:55:00  2018Q2  2018-04  249.86   \n",
       "\n",
       "   price_log  price_log_pre  log_return  jump  \n",
       "1   5.532401       5.539694   -0.007293     0  \n",
       "2   5.519860       5.532401   -0.012541     0  \n",
       "3   5.516649       5.519860   -0.003210     0  \n",
       "4   5.512864       5.516649   -0.003785     0  \n",
       "5   5.520901       5.512864    0.008037     0  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "jumps.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Part 5 - DF statistics with jump"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "def DF_Statistics_Jump(data):\n",
    "    \n",
    "    n = data.shape[0]\n",
    "    \n",
    "    log_return = data['log_return']\n",
    "    log_price_pre = data['price_log_pre']\n",
    "    jump = data['jump']\n",
    "    \n",
    "    alpha = np.ones(n)  # alpha = 1\n",
    "    beta = np.arange(n)  # beta*t = [0, 1, 2, ...]\n",
    "    \n",
    "    X = np.column_stack((log_price_pre, jump, alpha))\n",
    "    results = sm.OLS(log_return, X).fit()\n",
    "    \n",
    "    return results.tvalues[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-2.5001733056153133"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "DF_Statistics_Jump(data = jumps)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Part 6 - DF Critical Value using Monte Carlo Simulation "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "def generate_ar(n, phis, sigma=1):\n",
    "    \n",
    "    \n",
    "    '''\n",
    "        \n",
    "    Generates an autoregressive series.\n",
    "\n",
    "    The code follows: y_{t} = \\phi_{1} y_{t-1} + \\phi_{2} y_{t-2} + ... + \\phi_{p} y_{t-p} + \\epsilon_{t}\n",
    "\n",
    "    Args:\n",
    "        n: length of the series.\n",
    "        phis: list of thetas, in the order \\phi_{1}, \\phi_{2}, ..., \\phi_{p}.\n",
    "        sigma: standard deviation of the innovations (optional).\n",
    "\n",
    "    Returns:\n",
    "        np.Array with the series.\n",
    "    \n",
    "    '''\n",
    "\n",
    "    p = len(phis)\n",
    "    adj_n = n + p  # We add q values because at the beginning we have no phis available.\n",
    "    e_series = np.random.normal(0, sigma, size=adj_n)  # Generating a white noise.\n",
    "\n",
    "    ar = [e_series[0]]  # We start the series with a random value\n",
    "    for i in range(1, adj_n):\n",
    "        visible_phis = phis[0:min(p, i)]  # At first, we only \"see\" some of the phis.\n",
    "        visible_series = ar[i - min(p, i):i]  # The same happens to the white noise.\n",
    "        reversed_phis = visible_phis[::-1]\n",
    "\n",
    "        # Main equation.\n",
    "        ar_t = e_series[i] + np.dot(reversed_phis, visible_series)\n",
    "\n",
    "        ar.append(ar_t)\n",
    "\n",
    "    ar = ar[p:]  # Dropping the first values that did not use all the phis.\n",
    "\n",
    "    return ar\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "def dfcv_simulation(data, rep, jump_flag = True ):\n",
    "    \n",
    "    '''\n",
    "    \n",
    "    Generate critical value of DF distribution by monte carlo simulation\n",
    "    \n",
    "    \n",
    "    '''\n",
    "    \n",
    "    np.random.seed(10)\n",
    "    \n",
    "    df = data\n",
    "    start = df.day.min()\n",
    "    end = df.day.max()\n",
    "    repetitions = rep\n",
    "    series_length = data.shape[0]\n",
    "    percentiles = [2.5, 97.5]\n",
    "    t_values = []\n",
    "    \n",
    "    for _ in range(repetitions):  \n",
    "        y = generate_ar(series_length, [1])\n",
    "        delta_y = np.diff(y)  # delta_y_{t} = y_{t+1} - y_{t}\n",
    "        alpha = np.ones(series_length)  # alpha = 1\n",
    "        \n",
    "        if jump_flag == True:\n",
    "            jumps_value = df['jump'].values\n",
    "            regressors = np.column_stack((y, jumps_value, alpha))[:-1]\n",
    "        else:\n",
    "            regressors = np.column_stack((y, alpha))[:-1]\n",
    "        \n",
    "        results = sm.OLS(delta_y, regressors).fit()\n",
    "        t_values.append(results.tvalues[0])\n",
    "\n",
    "    #index = pd.MultiIndex.from_product([[start], [end]])\n",
    "    #df_table = pd.DataFrame(columns=percentiles, index=index)\n",
    "    \n",
    "    res = []\n",
    "    \n",
    "    for percentile in percentiles:\n",
    "        critical_t_value = np.percentile(t_values, percentile)\n",
    "        res.append(critical_t_value)\n",
    "        # df_table.loc[(start, end), percentile] = critical_t_value\n",
    "        \n",
    "    return res # df_table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[-2.483517128434954, 0.5670834013039572]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# CV without jumps\n",
    "dfcv_simulation(final, rep = 10, jump_flag = False )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[-2.8452384776298536, -0.19542959472079327]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# CV with jumps\n",
    "dfcv_simulation(jumps, rep = 10, jump_flag = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Part 7 - Bubble detection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "def bubble_check(path = 'TSLA.csv'):\n",
    "    \n",
    "    '''\n",
    "    \n",
    "    Buble check\n",
    "    \n",
    "    \n",
    "    \n",
    "    \n",
    "    '''\n",
    "    \n",
    "    # 1. Load data\n",
    "    data = load_data(path)\n",
    "    \n",
    "    # 2. Generate dataframe for outputs\n",
    "    df_jump = pd.DataFrame()\n",
    "    results = pd.DataFrame({'Start': pd.Series([], dtype='<M8[ns]'),\n",
    "                            'End': pd.Series([], dtype='<M8[ns]'),\n",
    "                            'Count': pd.Series([], dtype='int'),\n",
    "                            'DF': pd.Series([], dtype='float'),\n",
    "                            'Left_CV': pd.Series([], dtype='float'),\n",
    "                            'Right_CV': pd.Series([], dtype='float'),\n",
    "                            'DF_J': pd.Series([], dtype='float'),\n",
    "                            'Left_CV_J': pd.Series([], dtype='float'),\n",
    "                            'Right_CV_J': pd.Series([], dtype='float'),\n",
    "                            'Jumps': pd.Series([], dtype='int')})\n",
    "    \n",
    "    \n",
    "    # 3. brek data into month\n",
    "    cohort = data.month.unique()\n",
    "    \n",
    "    \n",
    "    for i in range(len(cohort)):\n",
    "\n",
    "        df = break_data(data, period = cohort[i])\n",
    "        \n",
    "        count = df.shape[0]\n",
    "        start = df.day.min()\n",
    "        end = df.day.max()\n",
    "        \n",
    "        df_score = DF_Statistics(df)\n",
    "        CV = dfcv_simulation(df, rep = 10, jump_flag = False )\n",
    "        \n",
    "        jumps = jump_detection(df, returns = 'log_return', alpha = 0.75)\n",
    "        df_jump = df_jump.append(jumps)\n",
    "        df_score_jump = DF_Statistics_Jump(data = jumps)\n",
    "        CVJ = dfcv_simulation(jumps, rep = 10, jump_flag = True)\n",
    "        no_jumps = jumps.jump.sum()\n",
    "        \n",
    "        results.loc[i, 'Start'] = start\n",
    "        results.loc[i, 'End'] = end\n",
    "        results.loc[i, 'Count'] = count\n",
    "        results.loc[i, 'DF'] = round(df_score,3)\n",
    "        results.loc[i, 'Left_CV'] = round(CV[0],3)\n",
    "        results.loc[i, 'Right_CV'] = round(CV[1],3)\n",
    "        \n",
    "        results.loc[i, 'DF_J'] = round(df_score_jump,3)\n",
    "        results.loc[i, 'Left_CV_J'] = round(CVJ[0],3)\n",
    "        results.loc[i, 'Right_CV_J'] = round(CVJ[1],3)\n",
    "        results.loc[i, 'Jumps'] = no_jumps\n",
    "        results['bubble'] = (results.DF_J > results.Right_CV_J) & (results.DF <= results.Right_CV)\n",
    "        #results.to_csv('bubble_test.csv')\n",
    "        \n",
    "    # Visulization\n",
    "    plt.figure(figsize = (25,5))\n",
    "    plt.plot('Index', 'price', data=df_jump)\n",
    "    IS = df_jump[df_jump['jump'] == 1]\n",
    "    plt.plot(IS[\"Index\"], IS[[\"price\"]], \"ro\", markersize=1)\n",
    "    plt.savefig(\"jumps.png\")\n",
    "\n",
    "     \n",
    "    return results, df_jump"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "month\n",
       "2017-04    0.024499\n",
       "2017-05    0.019633\n",
       "2017-06    0.013241\n",
       "2017-07    0.018448\n",
       "2017-08    0.011564\n",
       "2017-09    0.015833\n",
       "2017-10    0.018423\n",
       "2017-11    0.015524\n",
       "2017-12    0.016000\n",
       "2018-01    0.013872\n",
       "2018-02    0.009852\n",
       "2018-03    0.015743\n",
       "2018-04    0.013973\n",
       "2018-05    0.016695\n",
       "2018-06    0.015078\n",
       "2018-07    0.015700\n",
       "2018-08    0.020601\n",
       "2018-09    0.019333\n",
       "2018-10    0.018172\n",
       "2018-11    0.013358\n",
       "2018-12    0.017473\n",
       "2019-01    0.016285\n",
       "2019-02    0.019081\n",
       "2019-03    0.018697\n",
       "2019-04    0.015682\n",
       "2019-05    0.014968\n",
       "2019-06    0.016466\n",
       "2019-07    0.016213\n",
       "2019-08    0.015571\n",
       "2019-09    0.020899\n",
       "2019-10    0.014868\n",
       "2019-11    0.021154\n",
       "2019-12    0.022961\n",
       "2020-01    0.020833\n",
       "2020-02    0.021378\n",
       "2020-03    0.016166\n",
       "2020-04    0.014475\n",
       "2020-05    0.013300\n",
       "2020-06    0.014968\n",
       "2020-07    0.019574\n",
       "2020-08    0.017502\n",
       "2020-09    0.015719\n",
       "2020-10    0.012090\n",
       "2020-11    0.012579\n",
       "2020-12    0.013833\n",
       "Freq: M, Name: jump, dtype: float64"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_jump.groupby('month').jump.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    71949\n",
       "1     1211\n",
       "Name: jump, dtype: int64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_jump.jump.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "month\n",
      "2015-09       3\n",
      "2015-10       7\n",
      "2015-11      27\n",
      "2015-12      19\n",
      "2016-01      25\n",
      "2016-02      73\n",
      "2016-03      84\n",
      "2016-04     102\n",
      "2016-05     129\n",
      "2016-06     128\n",
      "2016-07     104\n",
      "2016-08     124\n",
      "2016-09     108\n",
      "2016-10      83\n",
      "2016-11     115\n",
      "2016-12     167\n",
      "2017-01     165\n",
      "2017-02     179\n",
      "2017-03     178\n",
      "2017-04    1370\n",
      "Freq: M, Name: price, dtype: int64\n",
      "(75137, 6)\n",
      "(73317, 6)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1800x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "if __name__ == '__main__':\n",
    "    results,df_jump = bubble_check()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}