linear_model.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "97d0ecd6",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "ba016864",
   "metadata": {},
   "outputs": [],
   "source": [
    "X = pd.read_csv(\"X.csv\", index_col=0)\n",
    "y = pd.read_csv(\"y.csv\", index_col=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "7168d1ed",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "3f90e6fa",
   "metadata": {},
   "outputs": [],
   "source": [
    "# standardize the data.\n",
    "ss_X = StandardScaler()\n",
    "X_ss = ss_X.fit_transform(X)\n",
    "ss_y = StandardScaler()\n",
    "y_ss = ss_y.fit_transform(np.array(y).reshape(-1, 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "5435cb58",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_ss = pd.DataFrame(X_ss, columns=[\"alpha1\", \"alpha2\", \"alpha3\", \"alpha4\", \"alpha5\", \"alpha6\", \"alpha7\"])\n",
    "y_ss = pd.DataFrame(y_ss, columns=[\"return\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "f0ef9da8",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.linear_model import Ridge\n",
    "from sklearn.model_selection import cross_val_score\n",
    "import statsmodels.api as sm\n",
    "from sklearn.metrics import r2_score\n",
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "a4c7994f",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_ss = sm.add_constant(X_ss)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "2ec10722",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\\begin{center}\n",
      "\\begin{tabular}{lclc}\n",
      "\\toprule\n",
      "\\textbf{Dep. Variable:}    &      return      & \\textbf{  R-squared:         } &     0.059   \\\\\n",
      "\\textbf{Model:}            &       OLS        & \\textbf{  Adj. R-squared:    } &     0.054   \\\\\n",
      "\\textbf{Method:}           &  Least Squares   & \\textbf{  F-statistic:       } &     14.49   \\\\\n",
      "\\textbf{Date:}             & Thu, 13 Apr 2023 & \\textbf{  Prob (F-statistic):} &  2.13e-18   \\\\\n",
      "\\textbf{Time:}             &     21:07:18     & \\textbf{  Log-Likelihood:    } &   -2277.6   \\\\\n",
      "\\textbf{No. Observations:} &        1640      & \\textbf{  AIC:               } &     4571.   \\\\\n",
      "\\textbf{Df Residuals:}     &        1632      & \\textbf{  BIC:               } &     4614.   \\\\\n",
      "\\textbf{Df Model:}         &           7      & \\textbf{                     } &             \\\\\n",
      "\\textbf{Covariance Type:}  &    nonrobust     & \\textbf{                     } &             \\\\\n",
      "\\bottomrule\n",
      "\\end{tabular}\n",
      "\\begin{tabular}{lcccccc}\n",
      "                & \\textbf{coef} & \\textbf{std err} & \\textbf{t} & \\textbf{P$> |$t$|$} & \\textbf{[0.025} & \\textbf{0.975]}  \\\\\n",
      "\\midrule\n",
      "\\textbf{const}  &    4.982e-15  &        0.024     &  2.07e-13  &         1.000        &       -0.047    &        0.047     \\\\\n",
      "\\textbf{alpha1} &       0.1395  &        0.037     &     3.813  &         0.000        &        0.068    &        0.211     \\\\\n",
      "\\textbf{alpha2} &       0.0632  &        0.026     &     2.472  &         0.014        &        0.013    &        0.113     \\\\\n",
      "\\textbf{alpha3} &       0.0241  &        0.040     &     0.608  &         0.543        &       -0.054    &        0.102     \\\\\n",
      "\\textbf{alpha4} &      -0.0600  &        0.040     &    -1.520  &         0.129        &       -0.138    &        0.017     \\\\\n",
      "\\textbf{alpha5} &       0.1028  &        0.024     &     4.198  &         0.000        &        0.055    &        0.151     \\\\\n",
      "\\textbf{alpha6} &       0.1459  &        0.026     &     5.665  &         0.000        &        0.095    &        0.196     \\\\\n",
      "\\textbf{alpha7} &       0.0663  &        0.025     &     2.626  &         0.009        &        0.017    &        0.116     \\\\\n",
      "\\bottomrule\n",
      "\\end{tabular}\n",
      "\\begin{tabular}{lclc}\n",
      "\\textbf{Omnibus:}       & 101.690 & \\textbf{  Durbin-Watson:     } &    2.003  \\\\\n",
      "\\textbf{Prob(Omnibus):} &   0.000 & \\textbf{  Jarque-Bera (JB):  } &  423.705  \\\\\n",
      "\\textbf{Skew:}          &  -0.076 & \\textbf{  Prob(JB):          } & 9.85e-93  \\\\\n",
      "\\textbf{Kurtosis:}      &   5.485 & \\textbf{  Cond. No.          } &     3.23  \\\\\n",
      "\\bottomrule\n",
      "\\end{tabular}\n",
      "%\\caption{OLS Regression Results}\n",
      "\\end{center}\n",
      "\n",
      "Notes: \\newline\n",
      " [1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n"
     ]
    }
   ],
   "source": [
    "regression1 = sm.OLS(y_ss, X_ss)\n",
    "model1 = regression1.fit()\n",
    "print(model1.summary().as_latex())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "0aa6d011",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train,X_test,y_train,y_test = train_test_split(X_ss, y_ss, random_state=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "f1a28706",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.020967364748314087"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reg_1 = LinearRegression()\n",
    "cross_val_score(reg_1,X_train,y_train,cv=10,scoring=\"r2\").mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "97996135",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.021045370329735823"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reg_2 = Ridge()\n",
    "cross_val_score(reg_2,X_train,y_train,cv=10,scoring=\"r2\").mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "051995d2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "linear reg of 1, 2, 5, 6, 7: 0.02301672427984224\n",
      "ridge reg of 1, 2, 5, 6, 7: 0.023077314838793307\n",
      "linear reg of 2, 3, 5, 6, 7: 0.01987591299283118\n",
      "ridge reg of 2, 3, 5, 6, 7: 0.019936738807049537\n",
      "linear reg of 2, 4, 5, 6, 7: 0.01834074905605738\n",
      "ridge reg of 2, 4, 5, 6, 7: 0.01840065375168278\n"
     ]
    }
   ],
   "source": [
    "# consider the correlation of alpha1, alpha3, alpha4, I take one of them and train model one by one.\n",
    "X_2 = X_train.iloc[:, [0, 1, 2, 5, 6, 7]] \n",
    "X_3 = X_train.iloc[:, [0, 2, 3, 5, 6, 7]]\n",
    "X_4 = X_train.iloc[:, [0, 2, 4, 5, 6, 7]]\n",
    "print(\"linear reg of 1, 2, 5, 6, 7: {}\".format(cross_val_score(reg_1,X_2,y_train,cv=10,scoring=\"r2\").mean()))\n",
    "print(\"ridge reg of 1, 2, 5, 6, 7: {}\".format(cross_val_score(reg_2,X_2,y_train,cv=10,scoring=\"r2\").mean()))\n",
    "print(\"linear reg of 2, 3, 5, 6, 7: {}\".format(cross_val_score(reg_1,X_3,y_train,cv=10,scoring=\"r2\").mean()))\n",
    "print(\"ridge reg of 2, 3, 5, 6, 7: {}\".format(cross_val_score(reg_2,X_3,y_train,cv=10,scoring=\"r2\").mean()))\n",
    "print(\"linear reg of 2, 4, 5, 6, 7: {}\".format(cross_val_score(reg_1,X_4,y_train,cv=10,scoring=\"r2\").mean()))\n",
    "print(\"ridge reg of 2, 4, 5, 6, 7: {}\".format(cross_val_score(reg_2,X_4,y_train,cv=10,scoring=\"r2\").mean()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "bd394ea7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0635398321255557"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reg_2.fit(X_test.iloc[:, [0, 1, 2, 5, 6, 7]],y_test)\n",
    "y_pre = reg_2.predict(X_test.iloc[:, [0, 1, 2, 5, 6, 7]])\n",
    "r2_score(y_test, y_pre)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "c1c8d081",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Price')"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_pre = ss_y.inverse_transform(reg_2.predict(X_test.iloc[:, [0, 1, 2, 5, 6, 7]]).reshape(-1, 1))\n",
    "cash = [[1], [1]]\n",
    "for i in range(len(y_test)):\n",
    "    cash[0].append(cash[0][-1]*(1 + ss_y.inverse_transform(np.array(y_test).reshape(-1, 1))[i]))\n",
    "    cash[1].append(cash[1][-1]*(1 + y_pre[i]))\n",
    "plt.plot(range(len(y_test)), cash[0][1:], label = \"true\")\n",
    "plt.plot(range(len(y_test)), cash[1][1:], label = \"random_forest\")\n",
    "plt.legend()\n",
    "plt.xlabel(\"Time\")\n",
    "plt.ylabel(\"Price\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "d9053aa8",
   "metadata": {},
   "outputs": [],
   "source": [
    "import joblib"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "301d575a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['saved_model/reg_ridge.pkl']"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "joblib.dump(reg_2, 'saved_model/reg_ridge.pkl')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "56e50feb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['saved_model/ss_y.pkl']"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "joblib.dump(ss_X, 'saved_model/ss_X.pkl')\n",
    "joblib.dump(ss_y, 'saved_model/ss_y.pkl')"
   ]
  }
 ],
 "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": "3.9.16"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}