Support & Resistance Line Algo v1 [Agglomerative Clustering].ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import yfinance as yf\n",
    "from sklearn.cluster import AgglomerativeClustering\n",
    "import mplfinance as mpf\n",
    "from datetime import date,timedelta\n",
    "\n",
    "# Calculate VERY simple waves\n",
    "df = yf.Ticker(\"SQ\").history(period=\"max\") # Download Data \n",
    "df.index = pd.to_datetime(df.index)\n",
    "df.drop(['Dividends','Stock Splits'], axis=1, inplace=True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1211 rows x 5 columns\n",
      " From Date 2015-11-19 to Date 2020-09-11 \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>Open</th>\n",
       "      <th>High</th>\n",
       "      <th>Low</th>\n",
       "      <th>Close</th>\n",
       "      <th>Volume</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2015-11-19</th>\n",
       "      <td>11.20</td>\n",
       "      <td>14.78</td>\n",
       "      <td>9.00</td>\n",
       "      <td>13.07</td>\n",
       "      <td>47466100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-09-11</th>\n",
       "      <td>143.91</td>\n",
       "      <td>143.91</td>\n",
       "      <td>134.26</td>\n",
       "      <td>137.45</td>\n",
       "      <td>8415600</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Open    High     Low   Close    Volume\n",
       "Date                                                \n",
       "2015-11-19   11.20   14.78    9.00   13.07  47466100\n",
       "2020-09-11  143.91  143.91  134.26  137.45   8415600"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def ends(df, n=2): \n",
    "    \"\"\" Credited to icfly2 : https://github.com/pandas-dev/pandas/issues/18691\n",
    "    Returns both head and tail of the dataframe or series. \n",
    "    Args:\n",
    "        n (int): Optional number of rows to return for each head and tail\n",
    "    \"\"\"\n",
    "    print('{} rows x {} columns'.format(np.shape(df)[0],np.shape(df)[1]))\n",
    "    print(' From Date {} to Date {} '.format(df.index[0].strftime('%Y-%m-%d'),df.index[-1].strftime('%Y-%m-%d')))\n",
    "    return df.head(n).append(df.tail(n))\n",
    "\n",
    "ends(df,n=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Credit to \n",
    "def generate_SR_line(df,rolling=20,start=False, end=False, duration=False, n=4, figsize = (20, 10), plot=True):\n",
    "    ''' \n",
    "    Credited to Mott The Tuple @ https://stackoverflow.com/questions/8587047/support-resistance-algorithm-technical-analysis/55311893#55311893\n",
    "    --------------------------------------------------------------------------------------------------------------------------\n",
    "    df = DataFrame \n",
    "    n = Number of Clusters / Number of Support/Resistance Line\n",
    "    Start = Start index  e.g. 0 \n",
    "    End = End index e.g. 100\n",
    "    Duration = # of days used in S&R line from today\n",
    "    --------------------------------------------------------------------------------------------------------------------------\n",
    "    '''\n",
    "    \n",
    "    # Choose Date Range\n",
    "    if duration == False:\n",
    "        df = df.iloc[start:end]\n",
    "    else:\n",
    "        df = df.iloc[-(duration+rolling):] # for the rolling obs waves \n",
    "    date = df.index\n",
    "    \n",
    "    # Reset_Index for merging \n",
    "    df.reset_index(inplace=True) \n",
    "\n",
    "    # Build rolling stats\n",
    "    mx = df.High.rolling(rolling).max().rename('waves')\n",
    "    mn = df.Low.rolling(rolling).min().rename('waves')\n",
    "\n",
    "    mx_waves = pd.concat([mx,pd.Series(np.zeros(len(mx))+1)],axis = 1)\n",
    "    mn_waves = pd.concat([mn,pd.Series(np.zeros(len(mn))+-1)],axis = 1)    \n",
    "\n",
    "    mx_waves.drop_duplicates('waves',inplace = True)\n",
    "    mn_waves.drop_duplicates('waves',inplace = True)\n",
    "\n",
    "    W = mx_waves.append(mn_waves).sort_index()\n",
    "    W = W[ W[0] != W[0].shift() ].dropna()\n",
    "\n",
    "    # Find Support/Resistance with clustering using the rolling stats.\n",
    "        # Create [x,y] array where y is always 1\n",
    "    X = np.concatenate((W.waves.values.reshape(-1,1),\n",
    "                        (np.zeros(len(W))+1).reshape(-1,1)), axis = 1 )\n",
    "    # Clustering Algo\n",
    "    cluster = AgglomerativeClustering(n_clusters=n,\n",
    "              affinity='euclidean', linkage='ward')\n",
    "    cluster.fit_predict(X)\n",
    "    W['clusters'] = cluster.labels_\n",
    "\n",
    "    # Get index of the max wave for each cluster\n",
    "    W2 = W.loc[W.groupby('clusters')['waves'].idxmax()]\n",
    "    df.index = date \n",
    "    # Plotting \n",
    "    if plot == True:\n",
    "        mpf.plot(df, type='candle', style='yahoo', volume=True,\n",
    "                 hlines=dict(hlines=W2.waves.to_list(), colors='g', linestyle='dashed', alpha=0.7),figsize=figsize)\n",
    "        \n",
    "    W2.waves.drop_duplicates(keep='first', inplace=True)\n",
    "    \n",
    "    return W2.reset_index().waves\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2000x1000 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "W2 = generate_SR_line(df, duration=100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Ending Date: 2016-04-15 00:00:00\n",
      "Ending Date: 2020-09-11 00:00:00\n"
     ]
    }
   ],
   "source": [
    "col_names = ['SR1','SR2','SR3','SR4']\n",
    "start = 0 \n",
    "period = 100 # roughly 1 quarter + 20 days for rolling data \n",
    "\n",
    "SRs = pd.DataFrame()\n",
    "for i in range(0, len(df) - period):\n",
    "    W = generate_SR_line(df, start = start+i, end = start+i+period, plot=False)\n",
    "    SRs = SRs.append(W.iloc[0:4], ignore_index=True)\n",
    "    if i == 0 or i == len(df) - period - 1:\n",
    "        print('Ending Date: ' + str(df.index[start+i+period]))\n",
    "SRs.columns = col_names\n",
    "SRs.index = df.iloc[period:].index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1111 rows x 9 columns\n",
      " From Date 2016-04-15 to Date 2020-09-11 \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>Open</th>\n",
       "      <th>High</th>\n",
       "      <th>Low</th>\n",
       "      <th>Close</th>\n",
       "      <th>Volume</th>\n",
       "      <th>SR1</th>\n",
       "      <th>SR2</th>\n",
       "      <th>SR3</th>\n",
       "      <th>SR4</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2016-04-15</th>\n",
       "      <td>14.85</td>\n",
       "      <td>14.87</td>\n",
       "      <td>14.18</td>\n",
       "      <td>14.33</td>\n",
       "      <td>2405100</td>\n",
       "      <td>15.91</td>\n",
       "      <td>10.90</td>\n",
       "      <td>9.63</td>\n",
       "      <td>11.80</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-04-18</th>\n",
       "      <td>14.20</td>\n",
       "      <td>14.71</td>\n",
       "      <td>14.13</td>\n",
       "      <td>14.51</td>\n",
       "      <td>1384700</td>\n",
       "      <td>12.90</td>\n",
       "      <td>10.90</td>\n",
       "      <td>10.80</td>\n",
       "      <td>15.13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-09-10</th>\n",
       "      <td>147.02</td>\n",
       "      <td>148.54</td>\n",
       "      <td>139.82</td>\n",
       "      <td>141.07</td>\n",
       "      <td>6711500</td>\n",
       "      <td>136.13</td>\n",
       "      <td>95.87</td>\n",
       "      <td>170.61</td>\n",
       "      <td>71.07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-09-11</th>\n",
       "      <td>143.91</td>\n",
       "      <td>143.91</td>\n",
       "      <td>134.26</td>\n",
       "      <td>137.45</td>\n",
       "      <td>8415600</td>\n",
       "      <td>136.13</td>\n",
       "      <td>95.87</td>\n",
       "      <td>170.61</td>\n",
       "      <td>71.07</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Open    High     Low   Close   Volume     SR1    SR2     SR3  \\\n",
       "Date                                                                         \n",
       "2016-04-15   14.85   14.87   14.18   14.33  2405100   15.91  10.90    9.63   \n",
       "2016-04-18   14.20   14.71   14.13   14.51  1384700   12.90  10.90   10.80   \n",
       "2020-09-10  147.02  148.54  139.82  141.07  6711500  136.13  95.87  170.61   \n",
       "2020-09-11  143.91  143.91  134.26  137.45  8415600  136.13  95.87  170.61   \n",
       "\n",
       "              SR4  \n",
       "Date               \n",
       "2016-04-15  11.80  \n",
       "2016-04-18  15.13  \n",
       "2020-09-10  71.07  \n",
       "2020-09-11  71.07  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dfs = pd.concat([df, SRs], axis=1).dropna()\n",
    "ends(dfs)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.7 (tensorflow)",
   "language": "python",
   "name": "tensorflow"
  },
  "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
}