Statistical_Arbitrage_Trading_System_1.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Statistical Arbitrage Trading System_1.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyPdo6GJL0VB13UErOl4CmrX",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/FinancialEngineerLab/finefinance/blob/master/Statistical_Arbitrage_Trading_System_1.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "moQwTpFa9pmB"
      },
      "source": [
        "Systematic Statistical Arbitrage Algorithmic Trading\n",
        "\n",
        "---\n",
        "\n",
        "\n",
        "Author : Shin Hyunjin (Quantitative Trader)\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wHcpZgua9iov",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "47105630-2d41-4da6-e887-34a8c37a65ba"
      },
      "source": [
        "# install library #\n",
        "!pip install -U finance_datareader\n",
        "!wget http://prdownloads.sourceforge.net/ta-lib/ta-lib-0.4.0-src.tar.gz\n",
        "!./configure --prefix=/usr\n",
        "!make\n",
        "!make install\n",
        "#!pip install Ta-Lib"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Collecting finance_datareader\n",
            "  Downloading https://files.pythonhosted.org/packages/ce/97/914f9c07de0edad6c1df84f0ee778b4fb5985d862eecf95a0ba3600ac109/finance_datareader-0.9.12-py3-none-any.whl\n",
            "Requirement already satisfied, skipping upgrade: lxml in /usr/local/lib/python3.6/dist-packages (from finance_datareader) (4.2.6)\n",
            "Requirement already satisfied, skipping upgrade: pandas>=0.19.2 in /usr/local/lib/python3.6/dist-packages (from finance_datareader) (1.1.5)\n",
            "Requirement already satisfied, skipping upgrade: tqdm in /usr/local/lib/python3.6/dist-packages (from finance_datareader) (4.41.1)\n",
            "Collecting requests-file\n",
            "  Downloading https://files.pythonhosted.org/packages/77/86/cdb5e8eaed90796aa83a6d9f75cfbd37af553c47a291cd47bc410ef9bdb2/requests_file-1.5.1-py2.py3-none-any.whl\n",
            "Requirement already satisfied, skipping upgrade: requests>=2.3.0 in /usr/local/lib/python3.6/dist-packages (from finance_datareader) (2.23.0)\n",
            "Requirement already satisfied, skipping upgrade: python-dateutil>=2.7.3 in /usr/local/lib/python3.6/dist-packages (from pandas>=0.19.2->finance_datareader) (2.8.1)\n",
            "Requirement already satisfied, skipping upgrade: pytz>=2017.2 in /usr/local/lib/python3.6/dist-packages (from pandas>=0.19.2->finance_datareader) (2018.9)\n",
            "Requirement already satisfied, skipping upgrade: numpy>=1.15.4 in /usr/local/lib/python3.6/dist-packages (from pandas>=0.19.2->finance_datareader) (1.19.4)\n",
            "Requirement already satisfied, skipping upgrade: six in /usr/local/lib/python3.6/dist-packages (from requests-file->finance_datareader) (1.15.0)\n",
            "Requirement already satisfied, skipping upgrade: urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1 in /usr/local/lib/python3.6/dist-packages (from requests>=2.3.0->finance_datareader) (1.24.3)\n",
            "Requirement already satisfied, skipping upgrade: certifi>=2017.4.17 in /usr/local/lib/python3.6/dist-packages (from requests>=2.3.0->finance_datareader) (2020.12.5)\n",
            "Requirement already satisfied, skipping upgrade: chardet<4,>=3.0.2 in /usr/local/lib/python3.6/dist-packages (from requests>=2.3.0->finance_datareader) (3.0.4)\n",
            "Requirement already satisfied, skipping upgrade: idna<3,>=2.5 in /usr/local/lib/python3.6/dist-packages (from requests>=2.3.0->finance_datareader) (2.10)\n",
            "Installing collected packages: requests-file, finance-datareader\n",
            "Successfully installed finance-datareader-0.9.12 requests-file-1.5.1\n",
            "--2020-12-23 23:23:09--  http://prdownloads.sourceforge.net/ta-lib/ta-lib-0.4.0-src.tar.gz\n",
            "Resolving prdownloads.sourceforge.net (prdownloads.sourceforge.net)... 216.105.38.13\n",
            "Connecting to prdownloads.sourceforge.net (prdownloads.sourceforge.net)|216.105.38.13|:80... connected.\n",
            "HTTP request sent, awaiting response... 301 Moved Permanently\n",
            "Location: http://downloads.sourceforge.net/project/ta-lib/ta-lib/0.4.0/ta-lib-0.4.0-src.tar.gz [following]\n",
            "--2020-12-23 23:23:09--  http://downloads.sourceforge.net/project/ta-lib/ta-lib/0.4.0/ta-lib-0.4.0-src.tar.gz\n",
            "Resolving downloads.sourceforge.net (downloads.sourceforge.net)... 216.105.38.13\n",
            "Reusing existing connection to prdownloads.sourceforge.net:80.\n",
            "HTTP request sent, awaiting response... 302 Found\n",
            "Location: https://managedway.dl.sourceforge.net/project/ta-lib/ta-lib/0.4.0/ta-lib-0.4.0-src.tar.gz [following]\n",
            "--2020-12-23 23:23:09--  https://managedway.dl.sourceforge.net/project/ta-lib/ta-lib/0.4.0/ta-lib-0.4.0-src.tar.gz\n",
            "Resolving managedway.dl.sourceforge.net (managedway.dl.sourceforge.net)... 208.79.213.6\n",
            "Connecting to managedway.dl.sourceforge.net (managedway.dl.sourceforge.net)|208.79.213.6|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 1330299 (1.3M) [application/x-gzip]\n",
            "Saving to: ‘ta-lib-0.4.0-src.tar.gz’\n",
            "\n",
            "ta-lib-0.4.0-src.ta 100%[===================>]   1.27M  7.59MB/s    in 0.2s    \n",
            "\n",
            "2020-12-23 23:23:10 (7.59 MB/s) - ‘ta-lib-0.4.0-src.tar.gz’ saved [1330299/1330299]\n",
            "\n",
            "/bin/bash: ./configure: No such file or directory\n",
            "make: *** No targets specified and no makefile found.  Stop.\n",
            "make: *** No rule to make target 'install'.  Stop.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_neNqIH39pVD"
      },
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import statsmodels\n",
        "import statsmodels.api as sm\n",
        "from statsmodels.tsa.stattools import coint, adfuller\n",
        "import matplotlib.pyplot as plt\n",
        "import FinanceDataReader as fdr\n",
        "import seaborn\n",
        "import datetime\n",
        "import scipy.stats as ss"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3HKq2Ey7Borc"
      },
      "source": [
        "\n",
        "---\n",
        "\n",
        "\n",
        "Data Frame Maker\n",
        "\n",
        "\n",
        "---\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "iQuhooERAvBH"
      },
      "source": [
        "## Data MGT ##\n",
        "def load_financial_data(ticker, start_date, end_date, output_file):\n",
        "  try:\n",
        "    df = pd.read_pickle(output_file)\n",
        "    print('File data found ... reading data')\n",
        "  except FileNotFoundError:\n",
        "    print('File not found ... downloading data')\n",
        "    df = pd.DataFrame(fdr.DataReader(ticker, start_date, end_date))\n",
        "    df.to_pickle(output_file)\n",
        "  return df\n",
        "\n",
        "def find_cointegrated_pairs(data):\n",
        "  n = data.shape[1]\n",
        "  score_matrix = np.zeros((n,n))\n",
        "  pvalue_matrix = np.ones((n,n))\n",
        "  keys = data.keys()\n",
        "  pairs = []\n",
        "  for i in range(n):\n",
        "    for j in range(i+1, n):\n",
        "      S1 = data[keys[i]]\n",
        "      S2 = data[keys[j]]\n",
        "      result = coint(S1, S2)\n",
        "      score = result[0]\n",
        "      pvalue = result[1]\n",
        "      score_matrix[i,j] = score\n",
        "      pvalue_matrix[i, j] = pvalue\n",
        "      if pvalue <= 0.02: # Significanc level\n",
        "        pairs.append((keys[i], keys[j]))\n",
        "  return score_matrix, pvalue_matrix, pairs\n",
        "\n",
        "def zscore(series):\n",
        "  return (series - series.mean())/ np.std(series)\n",
        "\n",
        "# basic concept of stationarity #\n",
        "def generate_data(params):\n",
        "  mu = params[0]\n",
        "  sigma = params[1]\n",
        "  return np.random.normal(mu, sigma)\n",
        "\n",
        "def stationarity_test(X, cutoff= 0.02):\n",
        "  pvalue = adfuller(X)[1]\n",
        "  if pvlalue < cutoff:\n",
        "    print('p-value = '+str(pvalue) + 'The Series '+ X.name + ' is likely stationary.')\n",
        "  else:\n",
        "    print('p-value = ' + str(pvalue) + 'The Series' + X.name +' is likely nonstationary.')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uQNKwQccA1Q8"
      },
      "source": [
        "### DataFrame ###\n",
        "ticker_list = [\"SPY\", \"JPY/KRW\", \"GBP/USD\",\"EUR/USD\",\n",
        "              \"JP225\",\"USTEC\",\"EU50\",\"HCEI\", \"KS200\", \"DE30\",\n",
        "               \"ZG\",\"HG\",\"T\", \"ZI\"] # hangseng no working\n",
        "              # HO : lean hog, GF :feeder Faddle, HE : lean hog, LE :live Cattle, ZC corn, ZL :soybean oil, RR : rice Oc1:oatmeal, zW: wheat\n",
        "symbolsids = ticker_list\n",
        "\n",
        "# time transformation #\n",
        "today = datetime.datetime.today()\n",
        "time1 = today + datetime.timedelta(days = -180)\n",
        "time2 = today + datetime.timedelta(days = -365)\n",
        "time3 = today + datetime.timedelta(days = -500)\n",
        "time4 = today + datetime.timedelta(days = -3000)\n",
        "time1 = datetime.datetime.strftime(time1,\"%Y-%m-%d\")\n",
        "time2 = datetime.datetime.strftime(time2,\"%Y-%m-%d\")\n",
        "time3 =datetime.datetime.strftime(time3,\"%Y-%m-%d\")\n",
        "time4 = datetime.datetime.strftime(time4,\"%Y-%m-%d\")\n",
        "\n",
        "## DataFrame Maker ##\n",
        "#import re\n",
        "#ticker_list = re.sub('[-=+,#‘|>`\\'…》]','',symbolsids)\n",
        "df_list_wtm = [fdr.DataReader(ticker, time1,)['Close'] for ticker in ticker_list]\n",
        "df_list_mtm = [fdr.DataReader(ticker, time2)['Close'] for ticker in ticker_list]\n",
        "df_list_year =[fdr.DataReader(ticker, time3)['Close'] for ticker in ticker_list]\n",
        "df_list_total =[fdr.DataReader(ticker, time4)['Close'] for ticker in ticker_list]\n",
        "\n",
        "#ZM , ZS, Platinum: soybean 제외\n",
        "#df_list_1 = [fdr.DataReader(ticker, \"2010-01-01\",\"2010-12-31\")['Close'] for ticker in ticker_list]\n",
        "#df_list_2 = [fdr.DataReader(ticker, \"2011-01-01\",\"2011-12-31\")['Close'] for ticker in ticker_list]\n",
        "#df_list_3 =[fdr.DataReader(ticker,  \"2012-01-01\",\"2012-12-31\")['Close'] for ticker in ticker_list]\n",
        "#df_list_4 =[fdr.DataReader(ticker,  \"2013-01-01\",\"2013-12-31\")['Close'] for ticker in ticker_list]\n",
        "#df_list_5 = [fdr.DataReader(ticker,  \"2014-01-01\",\"2014-12-31\")['Close'] for ticker in ticker_list]\n",
        "df_list_6 = [fdr.DataReader(ticker, \"2015-01-01\",\"2015-12-31\")['Close'] for ticker in ticker_list]\n",
        "df_list_7 =[fdr.DataReader(ticker,  \"2016-01-01\",\"2016-12-31\")['Close'] for ticker in ticker_list]\n",
        "df_list_8 =[fdr.DataReader(ticker, \"2017-01-01\",\"2017-12-31\")['Close'] for ticker in ticker_list]\n",
        "df_list_9 =[fdr.DataReader(ticker, \"2018-01-01\",\"2018-12-31\")['Close'] for ticker in ticker_list]\n",
        "df_list_10 =[fdr.DataReader(ticker, \"2019-01-01\",\"2019-12-31\")['Close'] for ticker in ticker_list]\n",
        "df_list_11 =[fdr.DataReader(ticker, \"2020-01-01\")['Close'] for ticker in ticker_list]\n",
        "\n",
        "#df_list_1 = pd.concat(df_list_1,axis=1)\n",
        "#df_list_2 = pd.concat(df_list_2,axis=1)\n",
        "#df_list_3 = pd.concat(df_list_3,axis=1)\n",
        "#df_list_4 = pd.concat(df_list_4,axis=1)\n",
        "#df_list_5 = pd.concat(df_list_5,axis=1)\n",
        "df_list_6 = pd.concat(df_list_6,axis=1)\n",
        "df_list_7 = pd.concat(df_list_7,axis=1)\n",
        "df_list_8 = pd.concat(df_list_8,axis=1)\n",
        "df_list_9 = pd.concat(df_list_9,axis=1)\n",
        "df_list_10 = pd.concat(df_list_10,axis=1)\n",
        "df_list_11 = pd.concat(df_list_11,axis=1)\n",
        "df_total = pd.concat([df_list_6,df_list_7,df_list_8,df_list_9,df_list_10, df_list_11])\n",
        "\n",
        "\n",
        "#df_list = pd.DataFrame(df_list)\n",
        "df1_raw = pd.concat(df_list_wtm, axis=1)\n",
        "df2_raw = pd.concat(df_list_mtm, axis=1)\n",
        "df3_raw = pd.concat(df_list_year, axis=1)\n",
        "df4_raw = df_total\n",
        "\n",
        "df1_raw.columns = ticker_list\n",
        "df1_raw = df1_raw.fillna(method = 'ffill')\n",
        "df2_raw.columns = ticker_list\n",
        "df2_raw = df2_raw.fillna(method = 'ffill')\n",
        "df3_raw.columns = ticker_list\n",
        "df3_raw = df3_raw.fillna(method = 'ffill')\n",
        "df4_raw.columns = ticker_list\n",
        "df4_raw = df4_raw.fillna(method = 'ffill')\n",
        "\n",
        "#df1 = np.log(df1_raw/df1_raw.shift(1))\n",
        "#df2 = np.log(df2_raw/df2_raw.shift(1))\n",
        "##df3 = np.log(df3_raw/df3_raw.shift(1))\n",
        "#df4 = np.log(df4_raw/df4_raw.shift(1))\n",
        "df1 = df1_raw\n",
        "df2 = df2_raw\n",
        "df3 = df3_raw\n",
        "df4 = df4_raw\n",
        "\n",
        "df1.columns = ticker_list\n",
        "df1 = df1.dropna()\n",
        "df2.columns = ticker_list\n",
        "df2 = df2.dropna()\n",
        "df3.columns = ticker_list\n",
        "df3 = df3.dropna()\n",
        "df4.columns = ticker_list\n",
        "df4 = df4.dropna()\n",
        "#print(df.head())\n",
        "#df\n",
        "\n",
        "score1, pvalues1, pairs1 = find_cointegrated_pairs(df1)\n",
        "score2, pvalues2, pairs2 = find_cointegrated_pairs(df2)\n",
        "score3, pvalues3, pairs3 = find_cointegrated_pairs(df3)\n",
        "score4, pvalues4, pairs4 = find_cointegrated_pairs(df4)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fya8itAj_iZm"
      },
      "source": [
        ""
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lusohC_fA1PM",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 318
        },
        "outputId": "7004d49e-cd83-4eb2-8fed-4379d94399cd"
      },
      "source": [
        "#pvalues visulization\n",
        "seaborn.heatmap(pvalues4, xticklabels = symbolsids,\n",
        "                yticklabels = symbolsids, cmap = 'RdYlGn_r',\n",
        "                mask = (pvalues4 >= 0.98)) # significance level"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7fda7fd83b38>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CqJwYdJxGr8l",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 102
        },
        "outputId": "8df900e2-2715-4d14-8111-25db5312a8a0"
      },
      "source": [
        "#print(pairs,len(pairs))\n",
        "result_pairs = list(set(pairs1).intersection(pairs2,pairs3,pairs4))\n",
        "result_pairs_2 = list(set(pairs2).intersection(pairs3))\n",
        "result_pairs_4= list(set(pairs4))\n",
        "result_pairs_3= list(set(pairs3))\n",
        "\n",
        "result_pairs_short = list(set(pairs1))\n",
        "result_pairs_mid = list(set(pairs2))\n",
        "\n",
        "pairs4\n",
        "#result_pairs_2"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('EUR/USD', 'KS200'),\n",
              " ('EUR/USD', 'HG'),\n",
              " ('EU50', 'HCEI'),\n",
              " ('KS200', 'HG'),\n",
              " ('DE30', 'HG')]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "w6VD9_mrGxhJ"
      },
      "source": [
        "\n",
        "\n",
        "---\n",
        "\n",
        "\n",
        "Spread Calculation\n",
        "\n",
        "\n",
        "---\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "c-IfBPwRGxZb",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 604
        },
        "outputId": "67337f43-a6bf-4f79-c1b8-b31dd03e3b2f"
      },
      "source": [
        "## Pair Selection Results ##\n",
        "\n",
        "# 주요 Pair : GOld copper, US10 -Bund10, Copper-US10, JP10Y - Yen\n",
        "# GBP - KS200\n",
        "# Palladium, Platinum\n",
        "\n",
        "ticker_result_1 = \"EUR/USD\"\n",
        "ticker_result_2 = \"KS200\"\n",
        "\n",
        "S1 = df4[ticker_result_1]\n",
        "S2 = df4[ticker_result_2]\n",
        "\n",
        "S1 = S1.dropna()\n",
        "S2 = S2.dropna()\n",
        "\n",
        "#S1 = pd.DataFrame(S1)\n",
        "#S2 = pd.DataFrame(S2)\n",
        "\n",
        "S1 = sm.add_constant(S1)\n",
        "results = sm.OLS(S2, S1).fit()\n",
        "S1 = S1[ticker_result_1]\n",
        "b = results.params[1]\n",
        "\n",
        "# spread #\n",
        "\n",
        "spread=  S2 - b * S1\n",
        "spread.plot(figsize = (12,6))\n",
        "plt.axhline(spread.mean(), color = 'black')\n",
        "plt.legend(['Spread'])\n",
        "plt.title('Pair Spread Results')\n",
        "plt.grid()\n",
        "\n",
        "df4[ticker_result_2]"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Date\n",
              "2015-01-05    244.26\n",
              "2015-01-06    239.93\n",
              "2015-01-07    240.53\n",
              "2015-01-08    243.94\n",
              "2015-01-09    246.05\n",
              "               ...  \n",
              "2020-09-23    309.64\n",
              "2020-09-24    302.48\n",
              "2020-09-25    303.57\n",
              "2020-09-26    303.57\n",
              "2020-09-27    303.57\n",
              "Name: KS200, Length: 1800, dtype: float64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1KsTIzYQ8i6a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 383
        },
        "outputId": "97265bdf-0454-4dbc-cd87-68bd7b1d5065"
      },
      "source": [
        "\n",
        "## Ratio Zscore ##\n",
        "#ratio =pd.DataFrame(ratio)\n",
        "\n",
        "zscore(spread).plot(figsize=  (12,6))\n",
        "plt.legend(['Spread Z score'])\n",
        "plt.title('Spread Pair Spread Z Score Trendlines')\n",
        "plt.axhline(2.0, color = 'red')\n",
        "plt.axhline(1.5, color = 'red')\n",
        "plt.axhline(1.0, color = 'red')\n",
        "plt.axhline(-1.0, color = 'blue')\n",
        "plt.axhline(-1.5, color = 'blue')\n",
        "plt.axhline(-2.0, color = 'blue')\n",
        "plt.axhline(0.0, color = 'black')\n",
        "plt.grid()\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "J9UroDIeoZcQ"
      },
      "source": [
        "## Pair Selection Results ##\n",
        "\n",
        "# 주요 Pair : GOld copper, US10 -Bund10, Copper-US10, JP10Y - Yen\n",
        "# GBP - KS200\n",
        "# Palladium, Platinum\n",
        "\n",
        "ticker_result_1 = \"GC\"\n",
        "ticker_result_2 = \"EUR/USD\"\n",
        "\n",
        "S1 = df4[ticker_result_1]\n",
        "S2 = df4[ticker_result_2]\n",
        "\n",
        "S1 = S1.dropna()\n",
        "S2 = S2.dropna()\n",
        "\n",
        "#S1 = pd.DataFrame(S1)\n",
        "#S2 = pd.DataFrame(S2)\n",
        "\n",
        "S1 = sm.add_constant(S1)\n",
        "results = sm.OLS(S2, S1).fit()\n",
        "S1 = S1[ticker_result_1]\n",
        "b = results.params[1]\n",
        "\n",
        "# spread #\n",
        "\n",
        "spread=  S2 - b * S1\n",
        "spread.plot(figsize = (12,6))\n",
        "plt.axhline(spread.mean(), color = 'black')\n",
        "plt.legend(['Spread'])\n",
        "plt.title('Pair Spread Results')\n",
        "plt.grid()\n",
        "\n",
        "df4[ticker_result_2]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "z2q_PF-noc5n"
      },
      "source": [
        "\n",
        "## Ratio Zscore ##\n",
        "#ratio =pd.DataFrame(ratio)\n",
        "\n",
        "zscore(spread).plot(figsize=  (12,6))\n",
        "plt.legend(['Spread Z score'])\n",
        "plt.title('Spread Pair Spread Z Score Trendlines')\n",
        "plt.axhline(2.0, color = 'red')\n",
        "plt.axhline(1.5, color = 'red')\n",
        "plt.axhline(1.0, color = 'red')\n",
        "plt.axhline(-1.0, color = 'blue')\n",
        "plt.axhline(-1.5, color = 'blue')\n",
        "plt.axhline(-2.0, color = 'blue')\n",
        "plt.axhline(0.0, color = 'black')\n",
        "plt.grid()\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gvnj4rpPT9tZ"
      },
      "source": [
        "pairs4[1][0]\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tnpJnga9NIMB"
      },
      "source": [
        "## Ratio ##\n",
        "\n",
        "ratio = S1/S2\n",
        "#ratio = pd.DataFrame(ratio)\n",
        "ratio = ratio.dropna()\n",
        "\n",
        "ratio.plot(figsize=  (12,6))\n",
        "plt.axhline(ratio.mean(), color = 'black')\n",
        "plt.legend(['Ratio Spread'])\n",
        "plt.title('Ratio Pair Spread Results')\n",
        "plt.grid()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3rboYp8A8pxn"
      },
      "source": [
        ""
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KaMnhMI19pXt"
      },
      "source": [
        "## Ratio Zscore ##\n",
        "#ratio =pd.DataFrame(ratio)\n",
        "ratio = ratio.dropna()\n",
        "\n",
        "zscore(ratio).plot(figsize=  (12,6))\n",
        "plt.legend(['Ratio Z score'])\n",
        "plt.title('Ratio Pair Spread Z Score Trendlines')\n",
        "plt.axhline(2.0, color = 'red')\n",
        "plt.axhline(1.5, color = 'red')\n",
        "plt.axhline(1.0, color = 'red')\n",
        "plt.axhline(-1.0, color = 'blue')\n",
        "plt.axhline(-1.5, color = 'blue')\n",
        "plt.axhline(-2.0, color = 'blue')\n",
        "plt.axhline(0.0, color = 'black')\n",
        "plt.grid()\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rDV7FJ3nbwUS"
      },
      "source": [
        "zscore(ratio).tail(30)\n",
        "#zscore(ratio).head()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "szeEY2imcXbH"
      },
      "source": [
        "## DataFrame Maker ##\n",
        "ticker_list = [ticker_result_1,ticker_result_2]\n",
        "df_list = [fdr.DataReader(ticker, time3)['Close'] for ticker in ticker_list]\n",
        "df_list\n",
        "\n",
        "df = pd.concat(df_list, axis=1)\n",
        "df.columns = ticker_list\n",
        "df = df.dropna()\n",
        "#df\n",
        "\n",
        "def corr_graph_result(ticker,ticker2, start_date):\n",
        "  corrdf = df[start_date:]\n",
        "\n",
        "  ax_0 = corrdf.plot(secondary_y = ticker2, figsize = (10,5))\n",
        "  plt.title('Graph : Two Ticker Graph')\n",
        "  plt.grid()\n",
        "  ax_0.get_legend().set_bbox_to_anchor((0.25, 0.85))\n",
        "  plt.show()\n",
        "\n",
        "corr_graph_result(ticker_result_1, ticker_result_2, time4)\n",
        "# place orders chart\n",
        "symbol_1 = df[ticker_result_1]\n",
        "symbol_2 = df[ticker_result_2]\n",
        "ratios = symbol_1 / symbol_2\n",
        "ratios.describe()\n",
        "ratios.plot()\n",
        "plt.title(\"  Ratio\")\n",
        "plt.axhline(ratios.mean())\n",
        "plt.legend(['Ratios'])\n",
        "plt.show()\n",
        "\n",
        "zscore = (ratios - ratios.mean())/ratios.std(ddof=0)\n",
        "zscore.plot()\n",
        "plt.title(\"Ratio Zscore (2Y)\")\n",
        "plt.axhline(zscore.mean(), color = \"black\")\n",
        "plt.axhline(1.0, color = \"red\")\n",
        "plt.axhline(-1.0, color = \"green\")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TRoCpvo7TEHx"
      },
      "source": [
        "ratio"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zoovfYbSOlrJ"
      },
      "source": [
        "\n",
        "\n",
        "---\n",
        "Trading Signals\n",
        "\n",
        "\n",
        "---\n",
        "\n",
        "* setup rules : Prediction Variable(Y) = sign(Ratio - Ratio[1])\n",
        "* Test : Train = 0.8 : 0.2\n",
        "\n",
        "\n",
        "\n",
        "---\n",
        "* Feature Engineering\n",
        " - 60 MA of ratio\n",
        " - 5 MA of ratio\n",
        " - 60 std\n",
        " - Z score\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eowpUmrP_dgx"
      },
      "source": [
        "# Test : Train\n",
        "#print(len(ratio)*0.7)\n",
        "train = ratio[:int(len(ratio)*0.7)]\n",
        "test = ratio[int(len(ratio)*0.7):]\n",
        "train"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rzKiUj92PDJA"
      },
      "source": [
        "#### Technical Analysis with Visulization ####\n",
        "\n",
        "ratio_mavg5 = train.rolling(window = 5, center = False).mean()\n",
        "ratio_mavg60 = train.rolling(window = 60, center = False).mean()\n",
        "std_60 = train.rolling(window = 60, center = False).std()\n",
        "zscore_60_5 = (ratio_mavg5 - ratio_mavg60)/std_60\n",
        "\n",
        "df_1 = pd.concat([train, ratio_mavg5, ratio_mavg60], axis =1)\n",
        "df_1.dropna()\n",
        "df_1.columns = ['train', 'ratio_mavg5', 'ratio_mavg60']\n",
        "df_1"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "78K0DkOnSopo"
      },
      "source": [
        "plt.figure(figsize= (12,6))\n",
        "plt.plot(df_1['train'])\n",
        "plt.plot(df_1['ratio_mavg5'])\n",
        "plt.plot(df_1['ratio_mavg60'])\n",
        "plt.legend(['Ratio', '5d Ratio MA', '60d Ratio MA'])\n",
        "plt.ylabel('Ratio')\n",
        "plt.title('Ratio, MA5, MA60')\n",
        "plt.grid()\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JRhHTOA4PDMG"
      },
      "source": [
        "#plt.legend(['Rolling Ratio Z score'])\n",
        "plt.figure(figsize = (20,15))\n",
        "plt.title('Rolling Ratio Pair Spread Z Score Trendlines')\n",
        "zscore_60_5.plot()\n",
        "plt.axhline(2.0, color = 'red', linestyle ='--')\n",
        "plt.axhline(1.5, color = 'red', linestyle ='--')\n",
        "plt.axhline(1.0, color = 'red', linestyle ='--')\n",
        "plt.axhline(0.0, color = 'black', linestyle ='--')\n",
        "plt.axhline(-1.0, color = 'blue', linestyle ='--')\n",
        "plt.axhline(-1.5, color = 'blue', linestyle ='--')\n",
        "plt.axhline(-2.0, color = 'blue', linestyle ='--')\n",
        "plt.legend(['Rolling Ratio Z score','+2','+1.5','+1.0','0','-1.0','-1.5','-2'])\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vgr5_Qttfuu6"
      },
      "source": [
        "\n",
        "\n",
        "---\n",
        "\n",
        "\n",
        "Modeling\n",
        "\n",
        "\n",
        "---\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "naILzU_pY1k2"
      },
      "source": [
        "plt.figure(figsize = (12,6))\n",
        "\n",
        "train[150:].plot()\n",
        "buy = train.copy()\n",
        "sell = train.copy()\n",
        "\n",
        "buy[zscore_60_5 > - 1] = 0\n",
        "sell[zscore_60_5 < 1] = 0\n",
        "buy[150:].plot(color ='g', linestyle = 'None', marker = '^')\n",
        "sell[150:].plot(color ='r', linestyle = 'None', marker = '^')\n",
        "\n",
        "x1, x2, y1, y2 = plt.axis()\n",
        "plt.axis((x1, x2, ratio.min(), ratio.max()))\n",
        "plt.legend(['Ratio',' Buy','Sell'])\n",
        "plt.grid()\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UPRQmHSsjjgA"
      },
      "source": [
        "plt.figure(figsize = (12,7))\n",
        "\n",
        "S1 = df4[ticker_result_1].iloc[:int(len(ratio)*0.7)]\n",
        "S2 = df4[ticker_result_2].iloc[:int(len(ratio)*0.7)]\n",
        "\n",
        "S1[60:].plot(color = 'b')\n",
        "S2[60:].plot(color = 'c')\n",
        "buyR = 0*S1.copy()\n",
        "sellR = 0 *S2.copy()\n",
        "\n",
        "# long ratio\n",
        "buyR[buy!=0] = S1[buy!=0]\n",
        "sellR[buy!=0] = S2[buy!=0]\n",
        "\n",
        "# short ratio\n",
        "buyR[sell!=0] = S2[sell!=0]\n",
        "sellR[sell!=0] = S2[sell!=0]\n",
        "\n",
        "# visulaization #\n",
        "buyR[60:].plot(color = 'g', linestyle = 'None', marker = '^')\n",
        "sellR[60:].plot(color = 'r', linestyle = 'None', marker = '^')\n",
        "#\n",
        "x1, x2, y1, y2 = plt.axis()\n",
        "plt.axis((x1, x2, min(S1.min(), S2.min()), max(S1.max(), S2.max())))\n",
        "plt.legend([ticker_result_1, ticker_result_2, 'Buy', 'Sell'])\n",
        "plt.grid()\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rWc9_qnoGZ56"
      },
      "source": [
        "\n",
        "\n",
        "---\n",
        "Trading System\n",
        "\n",
        "\n",
        "---\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kYIClo-vFhar"
      },
      "source": [
        "def trade(S1, S2, window1=60, window2=5):\n",
        "\n",
        "  #if window length is 0, algro not works\n",
        "  if (window1==0) or (window2 ==0 ):\n",
        "    return 0\n",
        "  \n",
        "  # mean and std\n",
        "  ratio= S1/S2\n",
        "  ma1 = ratio.rolling(window = window1, center = False).mean()\n",
        "  ma2 = ratio.rolling(window = window2, center = False).mean()\n",
        "  std = ratio.rolling(window = window2, center = False).std()\n",
        "  zscore= (ma1 - ma2)/std\n",
        "\n",
        "  # Money Management #\n",
        "  money = 0\n",
        "  countS1 = 0\n",
        "  countS2 = 0\n",
        "  for i in range(len(ratio)):\n",
        "    if zscore[i] < -1:\n",
        "      money += S1[i] - S2[i] * ratio[i]\n",
        "      countS1 -= 1\n",
        "      countS2 += ratio[i]\n",
        "    elif zscore[i] > 1:\n",
        "      money -= S1[i] - S2[i] * ratio[i]\n",
        "      countS1 += 1\n",
        "      countS2 -= ratio[i]\n",
        "    \n",
        "    #exit\n",
        "    elif abs(zscore[i]) < 0.75:\n",
        "      money+=S1[i] * countS1 + S2[i] * countS2\n",
        "      countS1 = 0\n",
        "      countS2 = 0\n",
        "  return money\n",
        "\n",
        "trade(df4[ticker_result_1].iloc[int(len(ratio)*0.7):], df4[ticker_result_2].iloc[int(len(ratio)*0.7):])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Oaqv_1ZhEBgO"
      },
      "source": [
        "'''\n",
        "#### initial set trailing stop for long position  \n",
        "context.longstop = data[stock].price * (1 - .05) # 5% stop-loss price\n",
        "\n",
        "# update: increase (never decrease) stop-loss  \n",
        "context.longstop = max( context.longstop, data[stock].price * (1 - .05) ) # 5% target\n",
        "\n",
        "# test stop-loss  \n",
        "if data[stock].price < context.longstop:  \n",
        "   # sell everything  '''"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "s9G-8bWBgCNv"
      },
      "source": [
        "\n",
        "## DataFrame Maker ##\n",
        "ticker_list = [ticker_result_1, ticker_result_2]\n",
        "df_list = [fdr.DataReader(ticker, time3)['Close'] for ticker in ticker_list]\n",
        "df_list\n",
        "\n",
        "df = pd.concat(df_list, axis=1)\n",
        "df.columns = ticker_list\n",
        "df = df.dropna()\n",
        "#df\n",
        "\n",
        "def corr_graph_result(ticker,ticker2, start_date):\n",
        "  corrdf = df[start_date:]\n",
        "\n",
        "  ax_0 = corrdf.plot(secondary_y = ticker2, figsize = (10,5))\n",
        "  plt.title('Graph : Two Ticker Graph')\n",
        "  plt.grid()\n",
        "  ax_0.get_legend().set_bbox_to_anchor((0.25, 0.85))\n",
        "  plt.show()\n",
        "\n",
        "corr_graph_result(ticker_result_1,ticker_result_2,time3)\n",
        "# place orders chart\n",
        "symbol_1 = df[ticker_result_1]\n",
        "symbol_2 = df[ticker_result_2]\n",
        "ratios = symbol_1 / symbol_2\n",
        "ratios.describe()\n",
        "ratios.plot()\n",
        "plt.title(\"Ratio\")\n",
        "plt.axhline(ratios.mean())\n",
        "plt.legend(['Ratios'])\n",
        "plt.show()\n",
        "\n",
        "zscore = (ratios - ratios.mean())/ratios.std(ddof=0)\n",
        "zscore.plot()\n",
        "plt.title(\"Ratio Zscore (1y)\")\n",
        "plt.axhline(zscore.mean(), color = \"black\")\n",
        "plt.axhline(1.0, color = \"red\")\n",
        "plt.axhline(-1.0, color = \"green\")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ggo5RTstgSXR"
      },
      "source": [
        "### Main Trading Strategy : Ratio Zscore ###\n",
        "\n",
        "# Strategy : Ratios(price1 / price2)'s Zscore\n",
        "# Enter Position : z score < -1 => short symbol1, long symbol2\n",
        "# Exit Position : next z score within -1 ~ +1 => long symbol1, short symbol2\n",
        "\n",
        "# Data Frame #\n",
        "symbol1_buy=symbol_1.copy()\n",
        "symbol1_sell=symbol_1.copy()\n",
        "symbol2_buy=symbol_2.copy()\n",
        "symbol2_sell=symbol_2.copy()\n",
        "symbol1_buy[zscore>-1] = 0\n",
        "symbol2_sell[zscore>-1] = 0\n",
        "symbol1_sell[zscore<1] = 0\n",
        "symbol2_buy[zscore<1] = 0\n",
        "\n",
        "pair_correlation_trading_strategy = pd.DataFrame(index = symbol_1.index)\n",
        "pair_correlation_trading_strategy['Symbol1_close'] = symbol_1\n",
        "pair_correlation_trading_strategy['Symbol2_close'] = symbol_2\n",
        "pair_correlation_trading_strategy['Symbol1_buy'] = np.zeros(len(symbol_1))\n",
        "pair_correlation_trading_strategy['Symbol1_sell'] = np.zeros(len(symbol_1))\n",
        "pair_correlation_trading_strategy['Symbol2_buy'] = np.zeros(len(symbol_1))\n",
        "pair_correlation_trading_strategy['Symbol2_sell'] = np.zeros(len(symbol_1))\n",
        "pair_correlation_trading_strategy['Delta'] = np.zeros(len(symbol_1))\n",
        "\n",
        "position = 0\n",
        "s1_shares = 1\n",
        "bigpointvalue1 = 100\n",
        "bigpointvalue2 = 50\n",
        "\n",
        "for i in np.arange(0, len(symbol_1)):\n",
        "  s1positions = symbol_1[i] * bigpointvalue1 * s1_shares\n",
        "  s2positions = symbol_2[i] * bigpointvalue2 * int(s1positions/(symbol_2[i]*bigpointvalue2)) \n",
        "  delta_position = s1positions-s2positions\n",
        "\n",
        "  if not position and symbol1_buy[i] != 0:\n",
        "    pair_correlation_trading_strategy['Symbol1_buy'][i] = s1positions\n",
        "    pair_correlation_trading_strategy['Symbol2_sell'][i] = s2positions\n",
        "    pair_correlation_trading_strategy['Delta'][i] = delta_position\n",
        "    position = 1\n",
        "  elif not position and symbol1_sell[i] != 0:\n",
        "    pair_correlation_trading_strategy['Symbol1_sell'][i] =  s1positions\n",
        "    pair_correlation_trading_strategy['Symbol2_buy'][i] =  s2positions\n",
        "    pair_correlation_trading_strategy['Delta'][i] = delta_position\n",
        "    position = -1\n",
        "  elif position == -1  and (symbol1_sell[i] ==0 or i == len(symbol_1)-1):\n",
        "    pair_correlation_trading_strategy['Symbol1_buy'][i] = s1positions\n",
        "    pair_correlation_trading_strategy['Symbol2_sell'][i] = s2positions\n",
        "    position = 0\n",
        "  elif position == 1 and  (symbol1_buy[i]==0 or i == len(symbol_2)-1):\n",
        "    pair_correlation_trading_strategy['Symbol1_sell'][i] =s1positions\n",
        "    pair_correlation_trading_strategy['Symbol2_buy'][i] =s2positions\n",
        "    position = 0\n",
        "\n",
        "pair_correlation_trading_strategy['Symbol1_position'] =pair_correlation_trading_strategy['Symbol1_buy']-pair_correlation_trading_strategy['Symbol1_sell']\n",
        "pair_correlation_trading_strategy['Symbol2_position'] =pair_correlation_trading_strategy['Symbol2_buy']-pair_correlation_trading_strategy['Symbol2_sell']\n",
        "pair_correlation_trading_strategy['Symbol1_position'].cumsum().plot(linestyle = \"--\")\n",
        "pair_correlation_trading_strategy['Symbol2_position'].cumsum().plot(linestyle = \":\")\n",
        "pair_correlation_trading_strategy['Total_position'] =pair_correlation_trading_strategy['Symbol1_position'] + pair_correlation_trading_strategy['Symbol2_position']\n",
        "pair_correlation_trading_strategy['Total_position'].cumsum().plot()\n",
        "plt.title(\"Symbol1 and Symbol2 positions\")\n",
        "plt.legend()\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "W23dLacCgTpN"
      },
      "source": [
        "pair_correlation_trading_strategy['Delta'].plot()\n",
        "plt.title(\"Delta Position\")\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}