{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Model Selection and Seasonal ARIMA\n",
    "\n",
    "Feng Li\n",
    "\n",
    "School of Statistics and Mathematics\n",
    "\n",
    "Central University of Finance and Economics\n",
    "\n",
    "[feng.li@cufe.edu.cn](mailto:feng.li@cufe.edu.cn)\n",
    "\n",
    "[https://feng.li/python](https://feng.li/python)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Estimation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Conditional Sum of Squares\n",
    "\n",
    "- A straightforward way to estimate ARMA models is via least squares. For an ARMA(1,1)\n",
    "\n",
    "$$\\underset{c,\\phi,\\theta}{\\min}\\sum_{t=1}^T\\left(y_t-c-\\phi_1 y_{t-1}-\\theta\\epsilon_{t-1}\\right)^2$$\n",
    "\n",
    "- Need to condition on initial values\n",
    "\n",
    "- Need to find a way to update all $\\epsilon$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Recursion\n",
    "\n",
    "- Use the following recursion to update the $\\epsilon$\n",
    "\n",
    "$$\\begin{aligned}\\color{blue}{\\epsilon_2}&=y_2-c-\\phi y_{1}-\\theta\\epsilon_1\\\\\\color{red}{\\epsilon_3}&=y_t-c-\\phi y_{2}-\\theta\\color{blue}{\\epsilon_2}\\\\\\epsilon_4&=y_t-c-\\phi y_{3}-\\theta\\color{red}{\\epsilon_3}\\\\\\vdots&=\\vdots\\quad\\vdots\\quad\\vdots\\end{aligned}$$\n",
    "\n",
    "- There is no closed form solution for $c$, $\\theta$ and $\\phi$ but they can be found numerically."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Initial values\n",
    "\n",
    "- There are many options but a common one is to condition on the first $p$ observed values of $y$ and set any $\\epsilon$ that need to be conditioned on to zero.\n",
    "- For an ARMA(1,1) condition on $y_1$ and set $\\epsilon_1=0$. Then use observations $t=2,\\dots,T$ in the sums of squares.\n",
    "\n",
    "$$\\underset{c,\\phi,\\theta}{\\min}\\sum_{t=2}^T\\left(y_t-c-\\phi_1 y_{t-1}-\\theta\\epsilon_{t-1}\\right)^2$$\n",
    "\n",
    "- Other options include setting $y_0=0$ or setting $y_0=E(Y)$ and using all of the data.\n",
    "- For stationary, invertible process, impact is minimal with enough data."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Likelihood\n",
    "\n",
    "- Maximum likelihood maximises the joint density of the data with respect to the parameters\n",
    "\n",
    "$$\\begin{aligned}\\hat{\\Theta},\\hat{\\Phi},\\hat\\sigma,\\hat c=&\\underset{c,\\Phi,\\Theta,\\sigma}{\\max}f(y_1,y_2,\\dots,y_T|c, \\Theta,\\Phi,\\sigma^2\\color{blue}{,y_0,y_{-1},\\dots,\\epsilon_0, \\epsilon_{-1},\\dots})\\\\=&\\underset{c,\\Phi,\\Theta,\\sigma}{\\max}L(c, \\Theta,\\Phi,\\sigma^2;y_1,y_2,\\dots,y_T\\color{blue}{,y_0,y_{-1},\\dots,\\epsilon_0, \\epsilon_{-1},\\dots})\\\\\\end{aligned}$$\n",
    "\n",
    "- Assume $\\epsilon_t\\sim N(0,\\sigma^2)$ for all $t$, i.e. assume normality\n",
    "- Conditional likelihood (includes parts in <span style=\"color:blue;\">blue</span>) equivalent to conditional sum of squares.\n",
    "- Unconditional likelihood does not have any of the parts in <span style=\"color:blue;\">blue</span> and requires the Kalman filter."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Advantages\n",
    "\n",
    "- Maximum likelihood is\n",
    "  - A consistent estimator\n",
    "  - Asymptotically normal\n",
    "  - Can be used to compute information criteria for model selection\n",
    "- Assuming normality is a disadvantage, but... \n",
    "- ...pseudo maximum likelihood theory establishes robustness to this assumption."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Information criteria\n",
    "\n",
    "For a model with $k$ parameters, all IC take the form\n",
    "\n",
    "$$IC = -2\\log L + k\\times q$$\n",
    "\n",
    "- The log likelihood measures the \"fit\" of the data, with higher values indicating a better fit.\n",
    "- To avoid overfitting there is a \"penalty\" on the number of parameters $k$.\n",
    "- If many ARIMA models are fit by maximum likelihood, information criteria can be computed for each model \n",
    "- Choose the model with the **lowest** value of the information criterion.\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## AIC, AICc and BIC\n",
    "\n",
    "- If $q=2$ we have the **AIC**. Theoretically the AIC will choose a model closest to the true data generating process (DGP), even if the true DGP is not one of the models estimated.\n",
    "- If $q=2+\\frac{2(k+1)}{T-k-1}$ we have the **AICc**. The AIC relies on asymptotic arguments, the AICc is a correction for finite samples.\n",
    "- if $q=log(T)$ we have the **BIC**. Theoretically the BIC is model consistent (it will choose the correct model as $T\\rightarrow\\infty$). However the true DGP must be one of the models under consideration.\n",
    "- In ARIMA modelling AICc is most commonly used"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Model Selection"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Auto Arima\n",
    "\n",
    "- Box-Jenkins approach of looking at ACF and PACF plots popular when computers were slow.\n",
    "- With improvements in computing it is possible to search through a large number of possible ARIMA models and select the best models using selection criteria.\n",
    "- One popular such algorithm is *auto arima*, proposed by [Hyndman and Khandakar (2008)](https://doi.org/10.18637/jss.v027.i03) and implemented in R packages.\n",
    "- For a Python implementation install [`statsforecast`](https://github.com/Nixtla/statsforecast)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Auto Arima (find $d$)\n",
    "\n",
    "- Step 1 is to find the correct level of differencing using a hypothesis test\n",
    "- By default the KPSS test is used.\n",
    "  - Null is that data are stationary\n",
    "  - If fail to reject, $d=0$. If rejected, take differences and apply KPSS test again. \n",
    "  - If fail to reject, $d=1$. If rejected a second time, $d=2$.\n",
    "- Other tests (augmented Dickey-Fuller and Phillips Perron) can be used.\n",
    "  - The null for these is that data are non-stationary."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Auto Arima (find $p,q$)\n",
    "\n",
    "- Fit four models\n",
    "  - ARIMA(0,d,0)\n",
    "  - ARIMA(1,d,0)\n",
    "  - ARIMA(0,d,1)\n",
    "  - ARIMA(2,d,2)\n",
    "- Select the model from the above list that minimises AICc"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "cell_style": "split",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Auto Arima (find $p,q$)\n",
    "\n",
    "Using the best current model search \"neighbouring models\", which are\n",
    "- Models with AR order different by $\\pm 1$.\n",
    "- Models with MA order different by $\\pm 1$.\n",
    "- Either one or both of these conditions makes the model a neighbouring model.\n",
    "- The algorithm moves to a better model as soon as it is found, but can also be forced to search all neighbouring models."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Model neighbourhood\n",
    "\n",
    "- What are the \"neghboring\" models of an ARMA(2,3)?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Application\n",
    "\n",
    "- Forecast turnover in restaurant/cafe takeway sector in NSW"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "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>Month</th>\n",
       "      <th>Turnover</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1982-04-01</td>\n",
       "      <td>85.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1982-05-01</td>\n",
       "      <td>84.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1982-06-01</td>\n",
       "      <td>80.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1982-07-01</td>\n",
       "      <td>82.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1982-08-01</td>\n",
       "      <td>80.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>436</th>\n",
       "      <td>2018-08-01</td>\n",
       "      <td>579.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>437</th>\n",
       "      <td>2018-09-01</td>\n",
       "      <td>569.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>438</th>\n",
       "      <td>2018-10-01</td>\n",
       "      <td>588.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>439</th>\n",
       "      <td>2018-11-01</td>\n",
       "      <td>576.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>440</th>\n",
       "      <td>2018-12-01</td>\n",
       "      <td>630.3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>441 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         Month  Turnover\n",
       "0   1982-04-01      85.4\n",
       "1   1982-05-01      84.8\n",
       "2   1982-06-01      80.7\n",
       "3   1982-07-01      82.4\n",
       "4   1982-08-01      80.7\n",
       "..         ...       ...\n",
       "436 2018-08-01     579.2\n",
       "437 2018-09-01     569.2\n",
       "438 2018-10-01     588.6\n",
       "439 2018-11-01     576.0\n",
       "440 2018-12-01     630.3\n",
       "\n",
       "[441 rows x 2 columns]"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "dat = pd.read_csv('data/takeaway.csv')\n",
    "dat['Month']=pd.to_datetime(dat['Month'])\n",
    "dat"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Time series plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f02777dd2b0>]"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1800x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "fig, ax = plt.subplots(figsize=(25, 6))\n",
    "ax.plot(dat.Month, dat.Turnover)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Transformation\n",
    "\n",
    "- Best to stabilise variance using log transformation\n",
    "- We can also split data into a test and training sample."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "dat['logTurnover'] = np.log(dat['Turnover'])\n",
    "train = dat.loc[1:404,:]\n",
    "test = dat.loc[405:,:]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Output\n",
    "\n",
    "- The `auto_arima_f` function finds the best model.\n",
    "  - The package is still under development and \n",
    "  - It is designed to work with many models and time series\n",
    "  - The classes created by the package are not easy to inspect compared to `statsmodels`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2, 3, 0, 0, 1, 1, 0)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "from statsforecast.arima import auto_arima_f\n",
    "out = auto_arima_f(train['logTurnover'].to_numpy())\n",
    "out['arma']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    " - The first two numbers are $p$ and $q$, sixth number is $d$, other numbers relate to seasonal ARIMA\n",
    " - The selected model is ARIMA(2,1,3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'coef': {'ar1': -1.3971657461229432,\n",
       "  'ar2': -0.5504300399710436,\n",
       "  'ma1': 1.1684981191760229,\n",
       "  'ma2': -0.04014085164881026,\n",
       "  'ma3': -0.30891348747978087},\n",
       " 'sigma2': 0.004209566597895503,\n",
       " 'var_coef': array([[5.23704348e-07, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
       "         0.00000000e+00],\n",
       "        [3.82730285e-06, 2.99479269e-06, 0.00000000e+00, 0.00000000e+00,\n",
       "         0.00000000e+00],\n",
       "        [0.00000000e+00, 0.00000000e+00, 6.15729424e-06, 0.00000000e+00,\n",
       "         0.00000000e+00],\n",
       "        [0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 6.15729424e-06,\n",
       "         0.00000000e+00],\n",
       "        [0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
       "         6.15729424e-06]]),\n",
       " 'mask': array([ True,  True,  True,  True,  True]),\n",
       " 'loglik': 533.2699228793731,\n",
       " 'aic': -1054.5398457587462,\n",
       " 'arma': (2, 3, 0, 0, 1, 1, 0),\n",
       " 'residuals': array([ 4.44029291e-03, -4.54976580e-02,  9.56731817e-03, -2.85556777e-02,\n",
       "         1.97013798e-02,  5.17932328e-02,  2.82506441e-02,  1.16781948e-01,\n",
       "        -1.42899268e-02, -6.15979826e-02, -5.75645930e-02, -6.06537902e-02,\n",
       "        -1.06080594e-02, -2.78585870e-02,  5.94797524e-02,  7.42004642e-03,\n",
       "         1.62131052e-02, -6.29711577e-02,  3.36299463e-02,  6.07726011e-02,\n",
       "         1.23908437e-01, -4.74269220e-02,  6.41449341e-02, -7.85209939e-02,\n",
       "         6.13893783e-02, -9.03865966e-02,  3.80051045e-02, -2.25966458e-02,\n",
       "         2.14866108e-02,  5.91745843e-02,  4.20980755e-02,  8.72813702e-02,\n",
       "        -2.88517060e-02, -1.15671673e-01,  1.64569053e-02,  1.27694578e-03,\n",
       "         3.50817869e-02, -4.59425262e-02,  5.39035026e-02,  2.41930136e-02,\n",
       "         3.56076982e-02,  5.75008641e-02,  8.56381479e-02,  1.00955466e-01,\n",
       "        -5.11762109e-02, -1.21457590e-01,  1.67806339e-02,  2.86316189e-02,\n",
       "         5.36152494e-02, -5.22152497e-02,  6.19240860e-03,  1.85203037e-02,\n",
       "        -1.02053159e-03,  4.70671774e-02,  4.78557185e-02,  1.60955617e-01,\n",
       "        -9.61759791e-02, -5.23276647e-02, -6.03278786e-02,  1.98220008e-02,\n",
       "        -2.42677211e-02, -7.28709972e-04,  8.33153986e-02,  1.17778891e-02,\n",
       "         4.53135418e-02,  4.80550024e-03,  3.31026618e-02,  7.13224696e-02,\n",
       "         1.19501841e-02, -7.77003836e-02,  4.40335930e-02, -4.68521147e-02,\n",
       "         2.97100543e-02, -4.50413089e-02,  3.23344469e-02, -2.64559051e-02,\n",
       "        -6.50246833e-03,  1.09983666e-02,  1.43383877e-02,  7.98739532e-02,\n",
       "         3.10050887e-03, -9.54046488e-02,  1.00778407e-02, -4.59702530e-02,\n",
       "         2.62118232e-02,  6.20216370e-02,  3.75803669e-02,  3.61005058e-02,\n",
       "         9.59609681e-03,  5.95501126e-02,  2.94191763e-02,  1.26478031e-01,\n",
       "         6.84442323e-02, -7.69584443e-02,  3.17738702e-02, -1.26414819e-02,\n",
       "         2.30656651e-02, -3.19328810e-02, -7.46516128e-05, -2.15484624e-03,\n",
       "        -1.32391135e-01,  5.03995421e-02, -2.19148749e-02,  1.46150918e-01,\n",
       "        -7.19257013e-02, -7.43303967e-02, -6.72985707e-03,  2.03940558e-02,\n",
       "        -1.86275483e-02, -1.33360522e-02,  1.41837682e-02,  2.63253676e-02,\n",
       "        -2.40452680e-02,  5.04812883e-02,  1.16516525e-02,  1.14231192e-01,\n",
       "         1.11658609e-02, -4.56830306e-02,  5.53201845e-02, -2.33089170e-02,\n",
       "         1.69945376e-02, -9.45191618e-02, -3.23511671e-02, -4.98810127e-02,\n",
       "         2.21704665e-02,  1.65701702e-02,  1.17521517e-02,  3.29278391e-02,\n",
       "        -6.59907292e-02, -1.25564503e-01, -1.09356147e-01,  3.57187028e-02,\n",
       "        -7.18026719e-02,  5.89486319e-02,  2.27798547e-02,  7.76411155e-02,\n",
       "         4.26051642e-02,  9.86611730e-02,  2.25670475e-02,  7.57197395e-02,\n",
       "         8.73363586e-02, -1.41633340e-01,  1.26265073e-01, -1.68216835e-01,\n",
       "         4.28707454e-02, -8.64513301e-02,  9.59173711e-02, -4.10543539e-02,\n",
       "         6.28820617e-02, -8.48397753e-02,  1.32520208e-02,  7.16106343e-02,\n",
       "         7.64904879e-02, -9.42331861e-02,  8.44571961e-02, -4.96688520e-03,\n",
       "         3.01483603e-02, -5.75843817e-02,  1.07921898e-01,  1.43466683e-02,\n",
       "         4.51074594e-02,  1.10322289e-01,  6.36546070e-02,  1.18259582e-01,\n",
       "         6.06386849e-02, -4.94522521e-02,  3.14331868e-02, -2.59256422e-02,\n",
       "        -1.23326613e-02, -6.29051059e-02, -9.85407438e-03, -1.56696679e-03,\n",
       "        -3.16011655e-02,  6.50863953e-02, -1.89623843e-02,  8.50598888e-02,\n",
       "         8.76205421e-03, -1.14723132e-01,  5.85380596e-02, -4.03829204e-02,\n",
       "         4.07823194e-02, -8.71286513e-02,  5.93192138e-02, -2.37316163e-02,\n",
       "         5.07361784e-02,  1.93334121e-05,  3.96629297e-03,  5.49561069e-02,\n",
       "        -2.77789331e-02, -1.08175484e-01, -1.28522508e-02, -1.52940127e-01,\n",
       "        -6.08791241e-02, -4.58251895e-02,  3.52823857e-02, -9.42911932e-02,\n",
       "        -8.18510203e-03,  3.74715382e-02, -5.19007412e-03,  7.76639455e-02,\n",
       "        -2.59009735e-02, -1.24703935e-01,  3.90903660e-02, -4.65669403e-02,\n",
       "         2.94166768e-02, -4.48243000e-02, -1.74959413e-02,  1.59992034e-02,\n",
       "         4.31485218e-02,  2.26347269e-02,  6.75451620e-03,  8.52911110e-04,\n",
       "        -5.31324556e-02, -1.00045602e-01,  4.42689648e-02,  3.26911424e-02,\n",
       "         7.28965368e-03,  3.16154367e-02,  1.75380354e-01,  4.35255661e-02,\n",
       "         1.14910453e-01,  8.53524968e-02, -5.48295411e-02,  9.17508131e-02,\n",
       "        -1.79676591e-02, -6.25646803e-02,  6.88278397e-02, -1.41354144e-01,\n",
       "        -5.65601524e-02, -3.53098106e-02,  4.44229860e-02, -1.92105318e-02,\n",
       "         2.42464469e-02,  6.90066100e-02, -3.77261105e-02,  1.40445849e-01,\n",
       "         5.72197415e-02, -1.62128136e-01,  6.79075355e-02, -2.34019368e-02,\n",
       "         1.77224286e-03, -5.41464918e-02,  8.56273371e-02, -4.16402866e-02,\n",
       "         1.94085884e-02,  5.42892556e-02, -3.10420475e-02,  1.14645909e-01,\n",
       "        -6.52151548e-03, -1.72741247e-01,  2.86268504e-02,  1.04143569e-01,\n",
       "         6.43158368e-02, -2.76706356e-02,  7.08642985e-02,  1.57671003e-02,\n",
       "         1.30281469e-02,  7.49839870e-02, -2.43497330e-02,  1.30414851e-01,\n",
       "        -1.42912226e-02, -9.63395932e-02, -9.80161966e-03,  9.25846941e-02,\n",
       "         1.32765116e-02, -1.20051425e-02,  6.16388885e-02, -4.25800038e-02,\n",
       "         3.38116034e-02,  1.14558411e-02, -4.33306887e-02,  7.25887059e-02,\n",
       "        -6.04710960e-02, -9.75983946e-02,  4.73093099e-03,  1.48605084e-02,\n",
       "        -6.71766030e-03,  6.15159524e-03,  1.18792985e-01, -4.44372531e-02,\n",
       "         4.02524606e-02,  8.76457573e-02,  3.26859777e-02,  1.07678114e-01,\n",
       "        -7.29658834e-02, -1.33640617e-01,  2.09458650e-02, -1.60300374e-02,\n",
       "         3.90596184e-04, -3.09976696e-02,  1.07115378e-01, -1.26259641e-02,\n",
       "         1.97574645e-02, -6.52697577e-02,  9.87741080e-03,  4.76141105e-02,\n",
       "        -2.50984411e-02, -1.32505589e-01,  7.19788377e-02, -1.83197442e-02,\n",
       "         3.12961695e-02, -7.49866233e-03,  1.18648644e-01,  1.12216759e-02,\n",
       "         4.87925530e-03,  1.87268526e-02, -1.46617905e-03,  6.72422507e-02,\n",
       "        -7.46876009e-02, -2.58371926e-02,  9.99855357e-03,  1.41942320e-02,\n",
       "         1.23944845e-02, -2.80792216e-02,  4.29071524e-02, -1.22531280e-02,\n",
       "        -4.18955786e-02,  6.95224956e-02, -3.79122985e-02,  1.72880281e-01,\n",
       "        -2.46544818e-02, -1.09147029e-01,  3.53859929e-02,  2.34379122e-02,\n",
       "         3.72420520e-02, -5.44117007e-02,  1.16148369e-01,  2.86986838e-03,\n",
       "         7.05058610e-02,  5.28870692e-02,  1.84301643e-02,  1.40401743e-01,\n",
       "         4.58460315e-03, -1.35954620e-01, -6.89105457e-04,  5.54593818e-02,\n",
       "         1.20335091e-02,  1.85011159e-02,  9.47188235e-02, -1.94164234e-02,\n",
       "         4.08086187e-02, -3.34565137e-03, -2.83909851e-02,  1.16481611e-01,\n",
       "        -8.65346211e-02, -1.58018014e-01, -2.54614621e-02, -3.99134183e-02,\n",
       "        -5.57000474e-02, -3.19554332e-02, -2.68952318e-02, -1.98500610e-02,\n",
       "         2.79791679e-02,  1.31351096e-03, -3.32015383e-02,  1.83497348e-01,\n",
       "        -5.38751213e-02, -1.17406732e-01,  7.63048335e-03,  7.95547995e-02,\n",
       "        -9.06377448e-03,  2.12562165e-02,  6.07248254e-02, -1.01372373e-02,\n",
       "         1.55207350e-02, -1.99711460e-02,  2.38260261e-02,  1.15180429e-01,\n",
       "        -1.17055915e-01, -1.34907749e-01,  7.20883687e-02,  1.00923052e-02,\n",
       "         8.58537220e-03, -2.00993491e-02,  1.21649692e-02,  3.76445983e-02,\n",
       "        -2.32873534e-02,  5.46469092e-02,  2.61834404e-02,  1.21012887e-01,\n",
       "        -6.13634722e-02, -1.06796596e-01,  5.90874766e-02,  1.43586220e-02,\n",
       "         3.21787239e-02, -5.01289045e-02,  9.16762265e-02,  1.08419307e-02,\n",
       "         5.24409858e-02,  3.28827014e-02,  4.91753653e-02,  1.13253857e-01,\n",
       "        -4.56721537e-02, -1.40014471e-01,  7.09532904e-02, -1.11604299e-02,\n",
       "         3.02575191e-02, -2.35080360e-02,  9.34882337e-02, -1.63134307e-02,\n",
       "         6.53457946e-02,  2.27425253e-02,  6.02643958e-03,  7.74224280e-02]),\n",
       " 'code': 2,\n",
       " 'n_cond': 0,\n",
       " 'nobs': 403,\n",
       " 'model': {'phi': array([-1.39716575, -0.55043004]),\n",
       "  'theta': array([ 1.16849812, -0.04014085, -0.30891349]),\n",
       "  'delta': array([1.]),\n",
       "  'Z': array([1., 0., 0., 0., 1.]),\n",
       "  'a': array([ 8.86840844e-02,  1.00895874e-01, -4.96945067e-03, -2.39168323e-02,\n",
       "          6.20010341e+00]),\n",
       "  'P': array([[ 2.22044605e-16,  2.22044605e-16,  2.77555756e-17,\n",
       "          -5.55111512e-17, -5.82486353e-17],\n",
       "         [ 2.22044605e-16,  2.44249065e-15,  5.13478149e-16,\n",
       "          -6.66133815e-16, -2.05101327e-16],\n",
       "         [ 2.77555756e-17,  5.13478149e-16,  1.44198889e-16,\n",
       "          -1.78676518e-16, -3.17634911e-17],\n",
       "         [-5.55111512e-17, -6.66133815e-16, -1.78676518e-16,\n",
       "           2.08166817e-16,  4.47052676e-17],\n",
       "         [-5.82486353e-17, -2.05101327e-16, -3.17634911e-17,\n",
       "           4.47052676e-17,  5.82486353e-17]]),\n",
       "  'T': array([[-1.39716575,  1.        ,  0.        ,  0.        ,  0.        ],\n",
       "         [-0.55043004,  0.        ,  1.        ,  0.        ,  0.        ],\n",
       "         [ 0.        ,  0.        ,  0.        ,  1.        ,  0.        ],\n",
       "         [ 0.        ,  0.        ,  0.        ,  0.        ,  0.        ],\n",
       "         [ 1.        ,  0.        ,  0.        ,  0.        ,  1.        ]]),\n",
       "  'V': array([[ 1.        ,  1.16849812, -0.04014085, -0.30891349,  0.        ],\n",
       "         [ 1.16849812,  1.36538785, -0.04690451, -0.36096483,  0.        ],\n",
       "         [-0.04014085, -0.04690451,  0.00161129,  0.01240005, -0.        ],\n",
       "         [-0.30891349, -0.36096483,  0.01240005,  0.09542754, -0.        ],\n",
       "         [ 0.        ,  0.        , -0.        , -0.        ,  0.        ]]),\n",
       "  'h': 0.0,\n",
       "  'Pn': array([[ 1.00000000e+00,  1.16849812e+00, -4.01408516e-02,\n",
       "          -3.08913487e-01,  8.64691236e-17],\n",
       "         [ 1.16849812e+00,  1.36538785e+00, -4.69045097e-02,\n",
       "          -3.60964829e-01, -3.59988979e-17],\n",
       "         [-4.01408516e-02, -4.69045097e-02,  1.61128797e-03,\n",
       "           1.24000505e-02, -3.75725852e-17],\n",
       "         [-3.08913487e-01, -3.60964829e-01,  1.24000505e-02,\n",
       "           9.54275427e-02,  0.00000000e+00],\n",
       "         [ 8.64691236e-17, -3.59988979e-17, -3.75725852e-17,\n",
       "           0.00000000e+00,  5.82486353e-17]])},\n",
       " 'bic': -1030.561133227032,\n",
       " 'aicc': -1054.327187530898,\n",
       " 'ic': None,\n",
       " 'xreg': None,\n",
       " 'x': array([4.44029554, 4.39073858, 4.41158544, 4.39073858, 4.40793802,\n",
       "        4.46935046, 4.47049528, 4.57677071, 4.53259949, 4.44382704,\n",
       "        4.42723898, 4.37826959, 4.39568296, 4.37826959, 4.4391156 ,\n",
       "        4.44500143, 4.43438187, 4.38327585, 4.41763506, 4.49535532,\n",
       "        4.58087749, 4.50313746, 4.54965748, 4.48751214, 4.53044664,\n",
       "        4.46935046, 4.48525989, 4.50313746, 4.49088104, 4.5716134 ,\n",
       "        4.58292458, 4.65014355, 4.60316818, 4.46590812, 4.53903038,\n",
       "        4.54648119, 4.56746832, 4.52396013, 4.5716134 , 4.60716819,\n",
       "        4.60716819, 4.66908351, 4.72650247, 4.79991426, 4.71133038,\n",
       "        4.58292458, 4.65681342, 4.69318106, 4.72561634, 4.66438205,\n",
       "        4.66626529, 4.70862889, 4.68490515, 4.7379513 , 4.77406872,\n",
       "        4.90970938, 4.77575649, 4.7022969 , 4.71133038, 4.72028299,\n",
       "        4.72028299, 4.70411013, 4.80402104, 4.78998862, 4.81055702,\n",
       "        4.82108769, 4.82831374, 4.90823336, 4.88507207, 4.79661665,\n",
       "        4.86368088, 4.822698  , 4.84024231, 4.81624116, 4.83310225,\n",
       "        4.82671246, 4.801559  , 4.83469334, 4.83786795, 4.91338991,\n",
       "        4.89858579, 4.7782828 , 4.82831374, 4.79330813, 4.81624116,\n",
       "        4.89485026, 4.89559848, 4.92071059, 4.91632461, 4.96633504,\n",
       "        4.9863426 , 5.086361  , 5.13226278, 5.00193085, 5.05879034,\n",
       "        5.05751888, 5.05879034, 5.04342512, 5.02978411, 5.04921478,\n",
       "        4.90453376, 4.99179221, 4.99247132, 5.10291057, 5.03239679,\n",
       "        4.91265489, 4.98292146, 4.98838969, 4.9705075 , 4.95864   ,\n",
       "        4.9781121 , 5.00662727, 4.96633504, 5.02388052, 5.02912988,\n",
       "        5.12336856, 5.11739483, 5.03304889, 5.12038616, 5.08450514,\n",
       "        5.08821343, 5.01196774, 4.98017609, 4.97742316, 4.99314998,\n",
       "        5.0271646 , 5.01661737, 5.05177724, 4.97535348, 4.85515039,\n",
       "        4.801559  , 4.87596039, 4.81462041, 4.86676492, 4.91118322,\n",
       "        4.94449549, 4.98770779, 5.04728861, 5.05241683, 5.09742442,\n",
       "        5.17614973, 4.99179221, 5.13990763, 4.99247132, 5.00193085,\n",
       "        5.00125813, 5.04213396, 5.04664573, 5.05560866, 5.00125813,\n",
       "        4.99653637, 5.10412564, 5.14224818, 5.01794187, 5.11978861,\n",
       "        5.11978861, 5.11379339, 5.07392303, 5.17388729, 5.18961795,\n",
       "        5.18794431, 5.31073989, 5.3264188 , 5.41119952, 5.44630624,\n",
       "        5.34758361, 5.40087395, 5.37481534, 5.35327892, 5.31172705,\n",
       "        5.30678144, 5.32981603, 5.28675073, 5.36550862, 5.33801873,\n",
       "        5.40312773, 5.4161004 , 5.26009615, 5.36877583, 5.33271879,\n",
       "        5.3499606 , 5.28978104, 5.33271879, 5.34233425, 5.35280555,\n",
       "        5.37481534, 5.34758361, 5.41743285, 5.37110304, 5.25332   ,\n",
       "        5.28826703, 5.14923714, 5.11859244, 5.13108145, 5.16192474,\n",
       "        5.07953927, 5.07204392, 5.14865659, 5.11379339, 5.19295685,\n",
       "        5.1550242 , 5.00796507, 5.10473262, 5.06259503, 5.08016136,\n",
       "        5.05815481, 5.02256386, 5.07267069, 5.09864617, 5.11259002,\n",
       "        5.10533923, 5.10230248, 5.05113724, 4.96004351, 5.04471461,\n",
       "        5.08140436, 5.05815481, 5.095589  , 5.25801607, 5.25540987,\n",
       "        5.31861007, 5.39544408, 5.2801534 , 5.38770093, 5.36597602,\n",
       "        5.2668267 , 5.38541207, 5.22143632, 5.18009674, 5.21112415,\n",
       "        5.23962837, 5.23431204, 5.23697374, 5.32056798, 5.25017699,\n",
       "        5.388615  , 5.4354672 , 5.20455599, 5.33416702, 5.3249593 ,\n",
       "        5.28826703, 5.27248661, 5.3442463 , 5.31271325, 5.30131288,\n",
       "        5.38678601, 5.31811999, 5.43720937, 5.420535  , 5.20290718,\n",
       "        5.30777253, 5.4275897 , 5.44068461, 5.39089655, 5.45702906,\n",
       "        5.47185042, 5.45189645, 5.53930084, 5.48604097, 5.60727157,\n",
       "        5.5831203 , 5.43938281, 5.491414  , 5.58724866, 5.57632782,\n",
       "        5.54165563, 5.6145869 , 5.55759996, 5.58236785, 5.61276308,\n",
       "        5.53851467, 5.63657372, 5.56298666, 5.45403824, 5.5174529 ,\n",
       "        5.53180721, 5.5174529 , 5.52585127, 5.64367855, 5.57063196,\n",
       "        5.58949333, 5.70311559, 5.68119598, 5.77919911, 5.67880649,\n",
       "        5.53180721, 5.62690143, 5.61130162, 5.60285656, 5.58687406,\n",
       "        5.68968355, 5.66850066, 5.65178718, 5.61203262, 5.613493  ,\n",
       "        5.68900719, 5.63300229, 5.50288953, 5.61895054, 5.60727157,\n",
       "        5.60763861, 5.62112524, 5.71636959, 5.71636959, 5.67948978,\n",
       "        5.71834263, 5.70111225, 5.76707021, 5.68255884, 5.64897424,\n",
       "        5.70444892, 5.69541422, 5.71406278, 5.67572588, 5.72423848,\n",
       "        5.71274222, 5.65284   , 5.75066606, 5.6957503 , 5.85736156,\n",
       "        5.81919233, 5.6503817 , 5.76268012, 5.77548279, 5.79270868,\n",
       "        5.73882782, 5.84845985, 5.85421195, 5.87183606, 5.94332417,\n",
       "        5.91025449, 6.05514332, 6.0224786 , 5.84643878, 5.90726739,\n",
       "        5.97787259, 5.96460709, 5.97482691, 6.06657202, 6.01956598,\n",
       "        6.04334517, 6.05185385, 5.99893656, 6.13902211, 6.02417372,\n",
       "        5.85248979, 5.91296232, 5.88638177, 5.83773045, 5.83276186,\n",
       "        5.8168135 , 5.80904299, 5.8444136 , 5.84238432, 5.79909265,\n",
       "        5.99670081, 5.90590666, 5.74652263, 5.83510309, 5.90889782,\n",
       "        5.87689509, 5.88610403, 5.95220383, 5.91754886, 5.9242558 ,\n",
       "        5.91377324, 5.92772571, 6.0530299 , 5.89302447, 5.7639364 ,\n",
       "        5.91593248, 5.9105256 , 5.89357605, 5.88749196, 5.89053863,\n",
       "        5.9396448 , 5.8957793 , 5.95220383, 5.97685839, 6.06796252,\n",
       "        5.98921201, 5.8576474 , 5.98418814, 5.98645201, 5.99321302,\n",
       "        5.95116325, 6.03356572, 6.05161847, 6.05795429, 6.10456999,\n",
       "        6.1180972 , 6.22673428, 6.14203741, 5.98745653, 6.12424566,\n",
       "        6.10969193, 6.11235386, 6.10702289, 6.18125812, 6.1649976 ,\n",
       "        6.19664777, 6.23225153, 6.20010341, 6.2887875 ]),\n",
       " 'lambda': None}"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "out"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Fit model using `statsmodels`\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>SARIMAX Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>      <td>logTurnover</td>   <th>  No. Observations:  </th>    <td>404</td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>            <td>ARIMA(2, 1, 3)</td>  <th>  Log Likelihood     </th>  <td>531.827</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>            <td>Thu, 07 Jul 2022</td> <th>  AIC                </th> <td>-1051.654</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                <td>13:28:47</td>     <th>  BIC                </th> <td>-1027.661</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Sample:</th>                  <td>0</td>        <th>  HQIC               </th> <td>-1042.155</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th></th>                      <td> - 404</td>      <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>        <td>opg</td>       <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "     <td></td>       <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ar.L1</th>  <td>    0.2047</td> <td>    3.820</td> <td>    0.054</td> <td> 0.957</td> <td>   -7.283</td> <td>    7.692</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ar.L2</th>  <td>    0.5400</td> <td>    2.431</td> <td>    0.222</td> <td> 0.824</td> <td>   -4.225</td> <td>    5.305</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.L1</th>  <td>   -0.5118</td> <td>    3.832</td> <td>   -0.134</td> <td> 0.894</td> <td>   -8.022</td> <td>    6.998</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.L2</th>  <td>   -0.6869</td> <td>    1.250</td> <td>   -0.549</td> <td> 0.583</td> <td>   -3.138</td> <td>    1.764</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.L3</th>  <td>    0.3407</td> <td>    1.572</td> <td>    0.217</td> <td> 0.828</td> <td>   -2.740</td> <td>    3.421</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>sigma2</th> <td>    0.0042</td> <td>    0.000</td> <td>   14.293</td> <td> 0.000</td> <td>    0.004</td> <td>    0.005</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Ljung-Box (L1) (Q):</th>     <td>0.59</td> <th>  Jarque-Bera (JB):  </th> <td>7.97</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Q):</th>                <td>0.44</td> <th>  Prob(JB):          </th> <td>0.02</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Heteroskedasticity (H):</th> <td>1.37</td> <th>  Skew:              </th> <td>-0.30</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(H) (two-sided):</th>    <td>0.07</td> <th>  Kurtosis:          </th> <td>3.33</td> \n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Covariance matrix calculated using the outer product of gradients (complex-step)."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                               SARIMAX Results                                \n",
       "==============================================================================\n",
       "Dep. Variable:            logTurnover   No. Observations:                  404\n",
       "Model:                 ARIMA(2, 1, 3)   Log Likelihood                 531.827\n",
       "Date:                Thu, 07 Jul 2022   AIC                          -1051.654\n",
       "Time:                        13:28:47   BIC                          -1027.661\n",
       "Sample:                             0   HQIC                         -1042.155\n",
       "                                - 404                                         \n",
       "Covariance Type:                  opg                                         \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "ar.L1          0.2047      3.820      0.054      0.957      -7.283       7.692\n",
       "ar.L2          0.5400      2.431      0.222      0.824      -4.225       5.305\n",
       "ma.L1         -0.5118      3.832     -0.134      0.894      -8.022       6.998\n",
       "ma.L2         -0.6869      1.250     -0.549      0.583      -3.138       1.764\n",
       "ma.L3          0.3407      1.572      0.217      0.828      -2.740       3.421\n",
       "sigma2         0.0042      0.000     14.293      0.000       0.004       0.005\n",
       "===================================================================================\n",
       "Ljung-Box (L1) (Q):                   0.59   Jarque-Bera (JB):                 7.97\n",
       "Prob(Q):                              0.44   Prob(JB):                         0.02\n",
       "Heteroskedasticity (H):               1.37   Skew:                            -0.30\n",
       "Prob(H) (two-sided):                  0.07   Kurtosis:                         3.33\n",
       "===================================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Covariance matrix calculated using the outer product of gradients (complex-step).\n",
       "\"\"\""
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import statsmodels as sm\n",
    "fitaa = sm.tsa.arima.model.ARIMA(train['logTurnover'], order = (2,1,3)).fit()\n",
    "fitaa.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Plot of forecast"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f0278ad0340>]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1800x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fcaa = fitaa.forecast(36)\n",
    "fig, ax = plt.subplots(1,figsize=(25,6))\n",
    "ax.plot(train['Month'].tail(48),train['logTurnover'].tail(48),color='black')\n",
    "ax.plot(test['Month'],test['logTurnover'],color='gray')\n",
    "ax.plot(test['Month'],fcaa,color='blue')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Seasonality\n",
    "\n",
    "- The ARIMA models as we have studied them so far do not take into account seasonality.\n",
    "- The forecasts work for short horizons, but fail to mimic the seasonal pattern at long horizons .\n",
    "- Looking at ACF and PACF plots leads to the same conclusion\n",
    "- Let's consider the ACF and PACF for the difference of log Turnover"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## ACF and PACF"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from statsmodels.graphics.tsaplots import plot_acf, plot_pacf\n",
    "fig, ax = plt.subplots(1,2,figsize=(15,6))\n",
    "dify = train.diff()['logTurnover'].iloc[2:]\n",
    "plot_acf(dify,ax=ax[0])\n",
    "plot_pacf(dify,ax=ax[1])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Pure Seasonal models\n",
    "\n",
    "- Seasonal AR(1): $Y_t=\\phi^{(s)} Y_{t-m}+\\epsilon_t$\n",
    "- Seasonal AR(2): $Y_t=\\phi_1^{(s)} Y_{t-m}+\\phi_2^{(s)} Y_{t-2m}+\\epsilon_t$\n",
    "- Seasonal AR(p): $(1-\\Phi^{(s)}(L^m))Y_t=\\epsilon_t$\n",
    "- Seasonal ARIMA(p,d,q): $(1-\\Phi^{(s)}(L^m))(1-L^m)^DY_t=(1+\\Theta^{(s)}(L^m))\\epsilon_t$\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Seasonal ARIMA\n",
    "\n",
    "The most general form for seasonal ARIMA is\n",
    "\n",
    "$$(1-\\Phi(L))(1-\\Phi^{(s)}(L^m))(1-L)^d(1-L^m)^DY_t=(1+\\Theta(L))(1+\\Theta^{(s)}(L^m))\\epsilon_t$$\n",
    "\n",
    "- Note that seasonal difference/components are always applied before non-seasonal components.\n",
    "\n",
    "- The auto arima algorithm can be generalised to seasonal ARIMA"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Auto arima (seasonal)\n",
    "\n",
    "Specify the seasonal period and set `seasonal=True` in `auto_arima_f`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "scrolled": false,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2, 3, 2, 2, 12, 0, 1)"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "out = auto_arima_f(train['logTurnover'].to_numpy(),seasonal=True,period=12)\n",
    "out['arma']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "- First two numbers are non-seasonal AR and MA orders.\n",
    "- Third and fourth numbers are seasonal AR and MA orders.\n",
    "- Fifth number is period.\n",
    "- Sixth and Seventh number are non-seasonal and seasonal orders of differencing."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['coef', 'sigma2', 'var_coef', 'mask', 'loglik', 'aic', 'arma', 'residuals', 'code', 'n_cond', 'nobs', 'model', 'xreg', 'bic', 'aicc', 'ic', 'x', 'lambda'])"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "out.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "scrolled": false,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'ar1': 1.495027198126436,\n",
       " 'ar2': -0.6042819420875728,\n",
       " 'ma1': -0.9348258615219661,\n",
       " 'ma2': 0.1883384110246554,\n",
       " 'ma3': 0.08816517296129588,\n",
       " 'sar1': -0.17736054462940312,\n",
       " 'sar2': 0.004977028288534635,\n",
       " 'sma1': -0.6182574588587337,\n",
       " 'sma2': -0.22384124235583505,\n",
       " 'drift': 0.004052622741266676}"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "out['coef']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[619.8529481890404, -1217.0093264572365, -1174.0501132182417]"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "[out['loglik'], out['aicc'], out['bic']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "scrolled": false,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 2.89379344e-07,  0.00000000e+00,  0.00000000e+00,\n",
       "         0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "         0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "         0.00000000e+00],\n",
       "       [-3.25946707e-06,  2.62594567e-06,  0.00000000e+00,\n",
       "         0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "         0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "         0.00000000e+00],\n",
       "       [ 0.00000000e+00,  0.00000000e+00,  6.50770512e-06,\n",
       "         0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "         0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "         0.00000000e+00],\n",
       "       [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "         6.50770512e-06,  0.00000000e+00,  0.00000000e+00,\n",
       "         0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "         0.00000000e+00],\n",
       "       [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "         0.00000000e+00,  6.50770512e-06,  0.00000000e+00,\n",
       "         0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "         0.00000000e+00],\n",
       "       [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "         0.00000000e+00,  0.00000000e+00,  6.50770512e-06,\n",
       "         0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "         0.00000000e+00],\n",
       "       [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "         0.00000000e+00,  0.00000000e+00,  2.31986211e-06,\n",
       "         6.50731361e-06,  0.00000000e+00,  0.00000000e+00,\n",
       "         0.00000000e+00],\n",
       "       [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "         0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "         0.00000000e+00,  6.50770512e-06,  0.00000000e+00,\n",
       "         0.00000000e+00],\n",
       "       [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "         0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "         0.00000000e+00,  0.00000000e+00,  6.50770512e-06,\n",
       "         0.00000000e+00],\n",
       "       [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "         0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "         0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "         6.50770512e-06]])"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "out['var_coef']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Plots"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>SARIMAX Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>              <td>logTurnover</td>            <th>  No. Observations:  </th>    <td>404</td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>           <td>ARIMA(2, 0, 3)x(2, 1, [1, 2], 12)</td> <th>  Log Likelihood     </th>  <td>669.876</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>                    <td>Thu, 07 Jul 2022</td>          <th>  AIC                </th> <td>-1319.751</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                        <td>13:39:13</td>              <th>  BIC                </th> <td>-1280.039</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Sample:</th>                          <td>0</td>                 <th>  HQIC               </th> <td>-1304.012</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th></th>                              <td> - 404</td>               <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>                <td>opg</td>                <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "      <td></td>        <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ar.L1</th>    <td>    1.5698</td> <td>    0.385</td> <td>    4.081</td> <td> 0.000</td> <td>    0.816</td> <td>    2.324</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ar.L2</th>    <td>   -0.5862</td> <td>    0.373</td> <td>   -1.572</td> <td> 0.116</td> <td>   -1.317</td> <td>    0.145</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.L1</th>    <td>   -0.7365</td> <td>    0.377</td> <td>   -1.952</td> <td> 0.051</td> <td>   -1.476</td> <td>    0.003</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.L2</th>    <td>    0.0039</td> <td>    0.072</td> <td>    0.055</td> <td> 0.956</td> <td>   -0.137</td> <td>    0.144</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.L3</th>    <td>    0.1126</td> <td>    0.048</td> <td>    2.332</td> <td> 0.020</td> <td>    0.018</td> <td>    0.207</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ar.S.L12</th> <td>   -0.4528</td> <td>    0.721</td> <td>   -0.628</td> <td> 0.530</td> <td>   -1.867</td> <td>    0.961</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ar.S.L24</th> <td>    0.0621</td> <td>    0.070</td> <td>    0.892</td> <td> 0.373</td> <td>   -0.074</td> <td>    0.199</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.S.L12</th> <td>   -0.3282</td> <td>    0.716</td> <td>   -0.459</td> <td> 0.646</td> <td>   -1.731</td> <td>    1.074</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.S.L24</th> <td>   -0.3748</td> <td>    0.537</td> <td>   -0.697</td> <td> 0.486</td> <td>   -1.428</td> <td>    0.678</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>sigma2</th>   <td>    0.0018</td> <td>    0.000</td> <td>   16.868</td> <td> 0.000</td> <td>    0.002</td> <td>    0.002</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Ljung-Box (L1) (Q):</th>     <td>2.01</td> <th>  Jarque-Bera (JB):  </th> <td>32.52</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Q):</th>                <td>0.16</td> <th>  Prob(JB):          </th> <td>0.00</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Heteroskedasticity (H):</th> <td>0.74</td> <th>  Skew:              </th> <td>-0.14</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(H) (two-sided):</th>    <td>0.08</td> <th>  Kurtosis:          </th> <td>4.38</td> \n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Covariance matrix calculated using the outer product of gradients (complex-step)."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                                       SARIMAX Results                                       \n",
       "=============================================================================================\n",
       "Dep. Variable:                           logTurnover   No. Observations:                  404\n",
       "Model:             ARIMA(2, 0, 3)x(2, 1, [1, 2], 12)   Log Likelihood                 669.876\n",
       "Date:                               Thu, 07 Jul 2022   AIC                          -1319.751\n",
       "Time:                                       13:39:13   BIC                          -1280.039\n",
       "Sample:                                            0   HQIC                         -1304.012\n",
       "                                               - 404                                         \n",
       "Covariance Type:                                 opg                                         \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "ar.L1          1.5698      0.385      4.081      0.000       0.816       2.324\n",
       "ar.L2         -0.5862      0.373     -1.572      0.116      -1.317       0.145\n",
       "ma.L1         -0.7365      0.377     -1.952      0.051      -1.476       0.003\n",
       "ma.L2          0.0039      0.072      0.055      0.956      -0.137       0.144\n",
       "ma.L3          0.1126      0.048      2.332      0.020       0.018       0.207\n",
       "ar.S.L12      -0.4528      0.721     -0.628      0.530      -1.867       0.961\n",
       "ar.S.L24       0.0621      0.070      0.892      0.373      -0.074       0.199\n",
       "ma.S.L12      -0.3282      0.716     -0.459      0.646      -1.731       1.074\n",
       "ma.S.L24      -0.3748      0.537     -0.697      0.486      -1.428       0.678\n",
       "sigma2         0.0018      0.000     16.868      0.000       0.002       0.002\n",
       "===================================================================================\n",
       "Ljung-Box (L1) (Q):                   2.01   Jarque-Bera (JB):                32.52\n",
       "Prob(Q):                              0.16   Prob(JB):                         0.00\n",
       "Heteroskedasticity (H):               0.74   Skew:                            -0.14\n",
       "Prob(H) (two-sided):                  0.08   Kurtosis:                         4.38\n",
       "===================================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Covariance matrix calculated using the outer product of gradients (complex-step).\n",
       "\"\"\""
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Fit a SARIMA model\n",
    "fitaas = sm.tsa.arima.model.ARIMA(train['logTurnover'], order = (2,0,3),seasonal_order=(2,1,2,12)).fit()\n",
    "fitaas.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f0278f44e80>]"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1800x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fcaas = fitaas.forecast(36)\n",
    "fig, ax = plt.subplots(1,figsize=(25,6))\n",
    "dat.Month, dat.Turnover\n",
    "ax.plot(train['Month'].tail(48),train['logTurnover'].tail(48),color='black')\n",
    "ax.plot(test['Month'],test['logTurnover'],color='gray')\n",
    "ax.plot(test['Month'],fcaas,color='blue')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Using Covariates\n",
    "\n",
    "- An important characteristic of (S)ARIMA models is the ability for them to include covariates or regressors.\n",
    "- The model becomes\n",
    "$$(1-\\Phi(L))(1-\\Phi^{(s)}(L^m))(1-L)^d(1-L^m)^D\\color{blue}{(y_t-\\mathbf{x}'_t\\boldsymbol{\\beta})}=(1+\\Theta(L))(1+\\Theta^{(s)}(L^m))\\epsilon_t$$\n",
    "\n",
    "- This is a regression model with (S)ARIMA errors."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "\n",
    "- If the purpose of the analysis is to forecast we can only use regressors that are themselves available when we make the forecast.\n",
    "- Consider forecasting demand for bike sharing scheme.\n",
    "- To forecast tomorrow's demand we cannot use tomorrow's weather.\n",
    "- We can however use dummy variables for the type of day, public holidays, etc."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Dataset\n",
    "\n",
    "DC bikeshare data of [Hadi and Gama (2013)](https://link.springer.com/article/10.1007/s13748-013-0040-3) from [UCI Machine Learning Repository](https://archive.ics.uci.edu/ml/datasets/bike%20sharing%20dataset)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f02787ee790>]"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1800x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "bikes=pd.read_csv('data/bike_sharing_daily.csv')\n",
    "bikes['dteday']=pd.to_datetime(bikes['dteday'])\n",
    "fig, ax = plt.subplots(1, figsize = (25, 6))\n",
    "ax.plot(bikes['dteday'], bikes['cnt'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Application\n",
    "\n",
    "- Can handle weekly seasonality by considering seasonal ARIMA models with a period of 7.\n",
    "- Handle yearly seasonality with month dummies\n",
    "- Keep holiday dummies\n",
    "- Convert data into numpy array as input to `auto_arima`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "bikes_clean = bikes[['cnt','mnth','holiday']]\n",
    "bikes_clean['int']=1.0\n",
    "bikes_clean = pd.get_dummies(bikes_clean, columns=['mnth'], drop_first=True)\n",
    "bikes_arr = bikes_clean.to_numpy(dtype='float')\n",
    "bikes_train=bikes_arr[:640,:]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Fit and forecast"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1, 1, 0, 1, 7, 1, 0)\n"
     ]
    }
   ],
   "source": [
    "out = auto_arima_f(bikes_train[:,0],xreg = bikes_train[:,1:],seasonal=True,period=7)\n",
    "print(out['arma'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "For this particular dataset auto arima chooses a SARIMA(1,1,1)(0,0,1)[7]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Fit model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>SARIMAX Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>                <td>y</td>              <th>  No. Observations:  </th>    <td>640</td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>           <td>ARIMA(1, 1, 1)x(0, 0, 1, 7)</td> <th>  Log Likelihood     </th> <td>-5229.620</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>                 <td>Thu, 07 Jul 2022</td>       <th>  AIC                </th> <td>10493.239</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                     <td>13:46:39</td>           <th>  BIC                </th> <td>10569.058</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Sample:</th>                       <td>0</td>              <th>  HQIC               </th> <td>10522.670</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th></th>                           <td> - 640</td>            <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>             <td>opg</td>             <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "     <td></td>        <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x1</th>      <td>  -24.5998</td> <td>  243.318</td> <td>   -0.101</td> <td> 0.919</td> <td> -501.495</td> <td>  452.296</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>const</th>   <td>    0.1058</td> <td> 9848.695</td> <td> 1.07e-05</td> <td> 1.000</td> <td>-1.93e+04</td> <td> 1.93e+04</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x2</th>      <td>  120.4969</td> <td>  765.482</td> <td>    0.157</td> <td> 0.875</td> <td>-1379.820</td> <td> 1620.813</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x3</th>      <td> 2043.6781</td> <td>  811.663</td> <td>    2.518</td> <td> 0.012</td> <td>  452.848</td> <td> 3634.508</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x4</th>      <td> 2370.7998</td> <td> 1079.991</td> <td>    2.195</td> <td> 0.028</td> <td>  254.056</td> <td> 4487.544</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x5</th>      <td> 1627.3375</td> <td> 1139.005</td> <td>    1.429</td> <td> 0.153</td> <td> -605.072</td> <td> 3859.747</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x6</th>      <td>  176.3790</td> <td> 1163.088</td> <td>    0.152</td> <td> 0.879</td> <td>-2103.232</td> <td> 2455.990</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x7</th>      <td>  168.3286</td> <td> 1258.711</td> <td>    0.134</td> <td> 0.894</td> <td>-2298.699</td> <td> 2635.356</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x8</th>      <td>  482.6837</td> <td> 1364.157</td> <td>    0.354</td> <td> 0.723</td> <td>-2191.014</td> <td> 3156.382</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x9</th>      <td>   57.8428</td> <td> 1402.966</td> <td>    0.041</td> <td> 0.967</td> <td>-2691.920</td> <td> 2807.605</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x10</th>     <td>-1229.7267</td> <td> 1442.379</td> <td>   -0.853</td> <td> 0.394</td> <td>-4056.737</td> <td> 1597.284</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x11</th>     <td> -529.5451</td> <td> 1248.308</td> <td>   -0.424</td> <td> 0.671</td> <td>-2976.184</td> <td> 1917.094</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x12</th>     <td> -113.0717</td> <td>  960.107</td> <td>   -0.118</td> <td> 0.906</td> <td>-1994.846</td> <td> 1768.703</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ar.L1</th>   <td>    0.3220</td> <td>    0.047</td> <td>    6.838</td> <td> 0.000</td> <td>    0.230</td> <td>    0.414</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.L1</th>   <td>   -0.8579</td> <td>    0.028</td> <td>  -30.259</td> <td> 0.000</td> <td>   -0.914</td> <td>   -0.802</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.S.L7</th> <td>   -0.0065</td> <td>    0.035</td> <td>   -0.185</td> <td> 0.853</td> <td>   -0.075</td> <td>    0.062</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>sigma2</th>  <td> 7.477e+05</td> <td> 2.73e+04</td> <td>   27.374</td> <td> 0.000</td> <td> 6.94e+05</td> <td> 8.01e+05</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Ljung-Box (L1) (Q):</th>     <td>0.00</td> <th>  Jarque-Bera (JB):  </th> <td>615.96</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Q):</th>                <td>0.96</td> <th>  Prob(JB):          </th>  <td>0.00</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Heteroskedasticity (H):</th> <td>2.76</td> <th>  Skew:              </th>  <td>-1.06</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(H) (two-sided):</th>    <td>0.00</td> <th>  Kurtosis:          </th>  <td>7.32</td> \n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Covariance matrix calculated using the outer product of gradients (complex-step)."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                                    SARIMAX Results                                    \n",
       "=======================================================================================\n",
       "Dep. Variable:                               y   No. Observations:                  640\n",
       "Model:             ARIMA(1, 1, 1)x(0, 0, 1, 7)   Log Likelihood               -5229.620\n",
       "Date:                         Thu, 07 Jul 2022   AIC                          10493.239\n",
       "Time:                                 13:46:39   BIC                          10569.058\n",
       "Sample:                                      0   HQIC                         10522.670\n",
       "                                         - 640                                         \n",
       "Covariance Type:                           opg                                         \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "x1           -24.5998    243.318     -0.101      0.919    -501.495     452.296\n",
       "const          0.1058   9848.695   1.07e-05      1.000   -1.93e+04    1.93e+04\n",
       "x2           120.4969    765.482      0.157      0.875   -1379.820    1620.813\n",
       "x3          2043.6781    811.663      2.518      0.012     452.848    3634.508\n",
       "x4          2370.7998   1079.991      2.195      0.028     254.056    4487.544\n",
       "x5          1627.3375   1139.005      1.429      0.153    -605.072    3859.747\n",
       "x6           176.3790   1163.088      0.152      0.879   -2103.232    2455.990\n",
       "x7           168.3286   1258.711      0.134      0.894   -2298.699    2635.356\n",
       "x8           482.6837   1364.157      0.354      0.723   -2191.014    3156.382\n",
       "x9            57.8428   1402.966      0.041      0.967   -2691.920    2807.605\n",
       "x10        -1229.7267   1442.379     -0.853      0.394   -4056.737    1597.284\n",
       "x11         -529.5451   1248.308     -0.424      0.671   -2976.184    1917.094\n",
       "x12         -113.0717    960.107     -0.118      0.906   -1994.846    1768.703\n",
       "ar.L1          0.3220      0.047      6.838      0.000       0.230       0.414\n",
       "ma.L1         -0.8579      0.028    -30.259      0.000      -0.914      -0.802\n",
       "ma.S.L7       -0.0065      0.035     -0.185      0.853      -0.075       0.062\n",
       "sigma2      7.477e+05   2.73e+04     27.374      0.000    6.94e+05    8.01e+05\n",
       "===================================================================================\n",
       "Ljung-Box (L1) (Q):                   0.00   Jarque-Bera (JB):               615.96\n",
       "Prob(Q):                              0.96   Prob(JB):                         0.00\n",
       "Heteroskedasticity (H):               2.76   Skew:                            -1.06\n",
       "Prob(H) (two-sided):                  0.00   Kurtosis:                         7.32\n",
       "===================================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Covariance matrix calculated using the outer product of gradients (complex-step).\n",
       "\"\"\""
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fitaasx = sm.tsa.arima.model.ARIMA(bikes_train[:,0],exog=bikes_train[:,1:],order = (1,1,1),seasonal_order=(0,0,1,7)).fit()\n",
    "fitaasx.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Forecast"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f02787dedf0>]"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1800x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "future_x =  bikes_arr[640:,1:]\n",
    "fc = fitaasx.forecast(len(future_x),exog = future_x)\n",
    "\n",
    "fig, ax = plt.subplots(1, figsize = (25, 6))\n",
    "ax.plot(bikes.iloc[550:640,1], bikes.iloc[550:640,15])\n",
    "ax.plot(bikes.iloc[640:,1], fc)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Fourier terms\n",
    "\n",
    "-  An alternative to month dummies is to use Fourier terms. For data with period $m$\n",
    "\n",
    "$$x_t^{(s)}=\\sin\\left(\\frac{2\\pi j t}{m}\\right)\\quad\\textrm{and}\\quad x_t^{(c)}=\\cos\\left(\\frac{2\\pi j t}{m}\\right)\\quad \\textrm{for}\\,j=1,2,\\dots,J$$\n",
    "\n",
    "- Fourier terms can be used to represent any periodic function to an arbitrary degree of accuracy.\n",
    "- In applied work, as few as 2-3 pairs of Fourier terms can be sufficient.\n",
    "- This is particularly useful for long seasonalities (e.g. $m=365$)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Wrap-up\n",
    "\n",
    "- Modern algorithms for selecting the order of ARIMA models are based on stepwise search, likelihood estimation and the AICc.\n",
    "- Seasonal ARIMA can capture seasonal effects with a small period.\n",
    "- Covariates based on calendar effects can be very useful for data based on human behaviour.\n",
    "- Regression with Fourier terms can capture seasonal effects with a long period. "
   ]
  }
 ],
 "metadata": {
  "celltoolbar": "Slideshow",
  "kernelspec": {
   "display_name": "Python3.9-virtualenv",
   "language": "python",
   "name": "python3.9-virtualenv"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.13"
  },
  "rise": {
   "autolaunch": true,
   "enable_chalkboard": true,
   "scroll": true,
   "start_slideshow_at": "beginning"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}