monkey-stocks.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Monkey Stocks\n",
    "\n",
    "O objetivo deste notebook é executar um backtesting de uma estratégia de seleção aleatória de portfolio baseada em [monkey stocks](https://www.instagram.com/monkeystocks/?hl=pt-br). A estratégia consiste, basicamente, em selecionar aleatoriamente 5 ativos da bolsa brasileira semanalmente para se manter comprado ao longo do periodo em questão.\n",
    "\n",
    "## Bibliotecas\n",
    "\n",
    "Foram necessários os seguintes módulos de python, para executar a estratégia.\n",
    "\n",
    "Inicialmente, é necessário obter os tickers das empresas listadas na [B3](http://www.b3.com.br/pt_br/), portanto, foi feito um crawler utilizando as bibliotecas [BeatifulSoup](https://www.crummy.com/software/BeautifulSoup/bs4/doc/) e [requests](https://requests.readthedocs.io/en/master/).\n",
    "\n",
    "Os dados de mercado foram obtidos da biblioteca [yfinance](https://aroussi.com/post/python-yahoo-finance) que busca os dados da API do [yahoo finance](https://finance.yahoo.com).\n",
    "\n",
    "A seleção aleatória da carteira de investimentos foi feita a partir do módulo [random](https://docs.python.org/2/library/random.html), interno de python.\n",
    "\n",
    "E as demais bibliotecas: numpy, pandas e matplotlib, foram utilizadas para visualização e manipulação de dados."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import yfinance as yf\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import random\n",
    "\n",
    "from bs4 import BeautifulSoup\n",
    "import requests"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Crawler dos Tickers da B3\n",
    "A função abaixo foi desenvolvida em nosso módulo [Quant-Utils](https://github.com/GrupoTuring/Quant-Utils), ela obtém a partir do site [Fundamentus](https://fundamentus.com.br) todos os tickers das empresas listadas na [B3](http://www.b3.com.br/pt_br/). Podendo ser filtrados por setor, conforme pode-se observar abaixo."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_tickers(setor=\"Todos\"):\n",
    "    '''\n",
    "    Essa função obtém os tickers listados no site fundamentus.com.br, seja um setor\n",
    "    específico, uma lista de setores ou todos os tickers de todos os setores.\n",
    "    '''\n",
    "    \n",
    "    setores = {\"Agropecuária\": \"42\", \"Água e Saneamento\": \"33\", \"Alimentos\": \"15\",\n",
    "               \"Bebidas\": \"16\", \"Comércio\": \"27\", \"Comércio2\": \"12\",\n",
    "               \"Comércio e Distribuição\": \"20\", \"Computadores e Equipamentos\": \"28\",\n",
    "               \"Construção e Engenharia\": \"13\", \"Diversos\": \"26\",\n",
    "               \"Embalagens\": \"6\", \"Energia Elétrica\": \"32\",\n",
    "               \"Equipamentos Elétricos\": \"9\", \"Exploração de Imóveis\": \"39\",\n",
    "               \"Financeiros\": \"35\", \"Fumo\": \"17\", \"Gás\": \"34\", \n",
    "               \"Holdings Diversificadas\": \"40\", \"Hoteis e Restaurantes\": \"24\",\n",
    "               \"Madeira e Papel\": \"5\", \"Máquinas e Equipamentos\": \"10\", \n",
    "               \"Materiais Diversos\": \"7\", \"Material de Transporte\": \"8\",\n",
    "               \"Mídia\": \"23\", \"Mineração\": \"2\", \"Outros\": \"41\",\n",
    "               \"Petróleo, Gás e Biocombustíveis\": \"1\", \"Previdência e Seguros\": \"38\",\n",
    "               \"Prods. de Uso Pessoal e de Limpeza\": \"18\", \"Programas e Serviços\": \"29\",\n",
    "               \"Químicos\": \"4\", \"Saúde\": \"19\", \"Securitizadoras de Recebíveis\": \"36\",\n",
    "               \"Serviços\": \"11\", \"Serviços Financeiros Diversos\": \"37\",\n",
    "               \"Siderurgia e Metalurgia\": \"3\", \"Tecidos, Vestuário e Calçados\": \"21\",\n",
    "               \"Telefonia Fixa\": \"30\", \"Telefonia Móvel\": \"31\", \"Transporte\": \"14\",\n",
    "               \"Utilidades Domésticas\": \"22\", \"Viagens e Lazer\": \"25\"}\n",
    "\n",
    "    lista_setores = []\n",
    "    if setor == \"Todos\":\n",
    "        lista_setores = list(setores.keys())\n",
    "    elif isinstance(setor, str):\n",
    "        if setores.has_key(setor):\n",
    "            lista_setores.append(ticker)\n",
    "        else:\n",
    "            print(\"Esse setor não existe\")\n",
    "            return []\n",
    "    elif isinstance(setor, list):\n",
    "        lista_setores = setor\n",
    "    else: \n",
    "        return []\n",
    "    \n",
    "    tickers = []\n",
    "\n",
    "    for item in lista_setores:\n",
    "        try:\n",
    "            url = \"https://www.fundamentus.com.br/resultado.php?setor=\" + setores[item]\n",
    "            response = requests.get(url)\n",
    "        except:\n",
    "            print(\"Não foi possível coletar informações do setor: \" + item)\n",
    "            continue\n",
    "\n",
    "        soup = BeautifulSoup(response.text, \"html.parser\")\n",
    "        tickers_bruto = soup.find('tbody').find_all(\"a\")\n",
    "        for ticker in tickers_bruto:\n",
    "            tickers.append(ticker.string + '.SA')\n",
    "        \n",
    "    tickers.sort()\n",
    "    return tickers"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Em seguida executamos a função ```get_tickers()``` que retorna uma lista de python com os tickers das empresas listadas.\n",
    "\n",
    "Exemplo:\n",
    "    - ITSA4.SA\n",
    "    - BIDI11.SA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# tickers e random tickers\n",
    "tickers = get_tickers()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Dados históricos de Mercado\n",
    "\n",
    "A função ```yf.download()``` retorna um pandas.DataFrame com os dados relacionados ao preço dos ativos para todos os ativos passados no parâmetro. Além disso, é necessário passar os parâmetros start e end, para dizer qual o periodo que se deseja obter os dados. Em caso de dúvidas vejam a [documentação](https://github.com/ranaroussi/yfinance).\n",
    "\n",
    "Note que nem todos os tickers estão disponíveis no periodo solicitado (talvez o ativo tenha sido retirado da bolsa ou ainda não esteja presente), nesses casos o pandas.DataFrame contém todos os valores NAN."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[*********************100%***********************]  402 of 402 completed\n",
      "\n",
      "4 Failed downloads:\n",
      "- LWSA3.SA: Data doesn't exist for startDate = 1546308000, endDate = 1577761200\n",
      "- MTRE3.SA: Data doesn't exist for startDate = 1546308000, endDate = 1577761200\n",
      "- PRNR3.SA: Data doesn't exist for startDate = 1546308000, endDate = 1577761200\n",
      "- MDNE3.SA: Data doesn't exist for startDate = 1546308000, endDate = 1577761200\n"
     ]
    }
   ],
   "source": [
    "start=\"2019-01-01\" # inicio do periodo a ser analisado\n",
    "end=\"2019-12-31\" # fim do periodo\n",
    "\n",
    "# dados historicos de mercado dos ativos (tickers)\n",
    "data = yf.download(tickers, start=start, end=end)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A função abaixo ```assets_return()``` calcula o retorno semanal dos tickers passados como parâmetro utilizando os preços de fechamento. E retorna um pandas.DataFrame contendo a soma dos retornos de cada ativo (sum) e média dos retornos (mean). "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def assets_return(data, start, end, tickers):\n",
    "    \n",
    "    df = data.Close[tickers].loc[start:end].pct_change().dropna()\n",
    "\n",
    "    df['sum'] = df.sum(axis=1)\n",
    "    df['mean'] = df.mean(axis=1)\n",
    "    return df[['sum', 'mean']]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A função principal desse notebook é a que segue ```monkey_stocks()``` que recebe os dados de preço de todos os ativos listados para o periodo analisado. E retorna um dataframe contendo os retornos semanais obtidos pela carteira.\n",
    "\n",
    "Para cada semana no periodo analisado, escolhemos aleatoriamente 5 ativos das empresas listadas na B3 e calculamos o seu retorno com base nos preços de fechamento. Em que o retorno ($r_t$) é performado pelo método ```pct_change()``` do pandas.DataFrame\n",
    "\n",
    "$$r_t = \\frac{P_t - P_{t-1}}{P_{t-1}}$$\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def monkey_stocks(data):\n",
    "\n",
    "    df = pd.DataFrame()\n",
    "    \n",
    "    # a cada semana\n",
    "    start2 = data.index[0]\n",
    "    for i in range(4, len(data.index), 4):\n",
    "        end2 = data.index[i]\n",
    "        \n",
    "        random_tickers = random.sample(tickers, 5)\n",
    "        \n",
    "        print(f\"Calculando semana {start} - {end}\")\n",
    "        print(f\"Tickers = {random_tickers}\")\n",
    "        \n",
    "        r = assets_return(data, start2, end2, random_tickers)\n",
    "        \n",
    "        df = pd.concat([df, r])\n",
    "        \n",
    "        start2 = end2\n",
    "        \n",
    "    return df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Abaixo temos a chamada da função que executa a estratégia. Performe mais de uma execução e visualize o resultado na seção de visualização."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['LAME4.SA', 'GEPA3.SA', 'GFSA3.SA', 'PNVL4.SA', 'AHEB3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['CPRE3.SA', 'AFLT3.SA', 'AHEB3.SA', 'BRAP4.SA', 'TRPL4.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['STBP3.SA', 'BBSE3.SA', 'DMMO3.SA', 'CVCB3.SA', 'GGBR3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['KLBN3.SA', 'CLSC4.SA', 'SULA11.SA', 'JFEN3.SA', 'BMGB4.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['MSPA3.SA', 'MYPK3.SA', 'MMXM3.SA', 'SPRI3.SA', 'MDIA3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['SAPR4.SA', 'CPLE3.SA', 'CEDO4.SA', 'PRNR3.SA', 'EKTR3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['BMEB4.SA', 'WLMM3.SA', 'BIDI4.SA', 'BRKM5.SA', 'BRKM3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['EALT4.SA', 'PATI4.SA', 'TUPY3.SA', 'PTNT4.SA', 'SHUL4.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['LPSB3.SA', 'KLBN11.SA', 'UNIP6.SA', 'VIVA3.SA', 'SULA3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['WLMM4.SA', 'TAEE4.SA', 'HYPE3.SA', 'LLIS3.SA', 'PATI4.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['CCPR3.SA', 'UNIP5.SA', 'HBOR3.SA', 'CEDO4.SA', 'PCAR3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['TXRX3.SA', 'TUPY3.SA', 'TCNO4.SA', 'GEPA4.SA', 'GEPA3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['TXRX4.SA', 'TCNO4.SA', 'HAGA4.SA', 'RSUL4.SA', 'SAPR4.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['EQTL3.SA', 'HAGA3.SA', 'WLMM3.SA', 'SANB4.SA', 'BRSR3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['SPRI6.SA', 'BRML3.SA', 'PRNR3.SA', 'INEP4.SA', 'CEDO3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['VIVT4.SA', 'USIM3.SA', 'OSXB3.SA', 'GOAU3.SA', 'AZEV3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['ETER3.SA', 'ABEV3.SA', 'PINE4.SA', 'SAPR4.SA', 'RADL3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['MRFG3.SA', 'CGAS5.SA', 'MEAL3.SA', 'MGLU3.SA', 'CARD3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['FESA3.SA', 'OIBR4.SA', 'CSRN3.SA', 'SBSP3.SA', 'EUCA4.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['BRKM6.SA', 'BRML3.SA', 'ALUP11.SA', 'IDVL3.SA', 'BALM4.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['CTNM3.SA', 'CVCB3.SA', 'PATI4.SA', 'JOPA3.SA', 'NORD3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['BBDC3.SA', 'BPAN4.SA', 'BMIN4.SA', 'MSPA3.SA', 'CRPG6.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['GEPA3.SA', 'ALPA4.SA', 'BKBR3.SA', 'SGPS3.SA', 'ITSA3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['PATI4.SA', 'BRDT3.SA', 'SQIA3.SA', 'TEKA3.SA', 'SLCE3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['CESP3.SA', 'BIOM3.SA', 'MNPR3.SA', 'VULC3.SA', 'EGIE3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['ENBR3.SA', 'ALUP4.SA', 'SANB4.SA', 'ECOR3.SA', 'CRPG6.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['SLED4.SA', 'CEBR5.SA', 'PEAB4.SA', 'RANI4.SA', 'CTSA3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['JOPA4.SA', 'HYPE3.SA', 'RCSL4.SA', 'EQTL3.SA', 'BEEF3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['CARD3.SA', 'RAIL3.SA', 'CPLE3.SA', 'BRAP4.SA', 'RANI3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['ENEV3.SA', 'CEBR6.SA', 'MNDL3.SA', 'VULC3.SA', 'RNEW3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['TASA4.SA', 'ENMT3.SA', 'PEAB3.SA', 'RANI4.SA', 'PSSA3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['CPLE3.SA', 'AZEV4.SA', 'DASA3.SA', 'FRAS3.SA', 'CPRE3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['CRFB3.SA', 'AZUL4.SA', 'CTNM3.SA', 'CEDO3.SA', 'SULA11.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['TCNO3.SA', 'TIET3.SA', 'SBSP3.SA', 'NORD3.SA', 'PETR3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['TPIS3.SA', 'BBAS3.SA', 'EZTC3.SA', 'ECOR3.SA', 'ADHM3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['CARD3.SA', 'AZEV4.SA', 'MRVE3.SA', 'WHRL4.SA', 'RANI3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['BAUH4.SA', 'BRDT3.SA', 'NEOE3.SA', 'ALSO3.SA', 'AHEB5.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['TRIS3.SA', 'TELB4.SA', 'TEKA3.SA', 'VIVT4.SA', 'NEOE3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['JBDU4.SA', 'VIVT3.SA', 'ENAT3.SA', 'MTIG4.SA', 'VIVT4.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['BEES4.SA', 'GEPA3.SA', 'CRFB3.SA', 'MTRE3.SA', 'WHRL3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['ALPA3.SA', 'CGRA4.SA', 'PMAM3.SA', 'WEGE3.SA', 'RPAD3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['TAEE3.SA', 'EQPA3.SA', 'TRPL3.SA', 'SGPS3.SA', 'WSON33.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['SEER3.SA', 'CSMG3.SA', 'UCAS3.SA', 'BRAP4.SA', 'VULC3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['SANB4.SA', 'CRIV4.SA', 'BEES3.SA', 'TECN3.SA', 'BBDC4.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['HAGA3.SA', 'BRGE11.SA', 'PINE4.SA', 'WLMM4.SA', 'NORD3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['GEPA3.SA', 'USIM5.SA', 'SLED4.SA', 'TUPY3.SA', 'CGRA3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['DTEX3.SA', 'ALUP11.SA', 'SPRI5.SA', 'OMGE3.SA', 'ENEV3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['WLMM4.SA', 'QUAL3.SA', 'ALSO3.SA', 'BRGE6.SA', 'TIET4.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['RNEW4.SA', 'LUPA3.SA', 'CRDE3.SA', 'BRSR5.SA', 'BRGE8.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['JBDU4.SA', 'IDVL4.SA', 'CARD3.SA', 'TRIS3.SA', 'AZEV4.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['GEPA4.SA', 'NORD3.SA', 'LPSB3.SA', 'EVEN3.SA', 'CSRN3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['BMEB3.SA', 'MDIA3.SA', 'PNVL4.SA', 'PSSA3.SA', 'TEKA4.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['CPLE6.SA', 'MNPR3.SA', 'KLBN4.SA', 'CPFE3.SA', 'RDNI3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['STBP3.SA', 'APER3.SA', 'MSPA3.SA', 'BMIN4.SA', 'LAME4.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['ELET3.SA', 'ALUP3.SA', 'MWET4.SA', 'TELB4.SA', 'TRPL4.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['PINE4.SA', 'INEP4.SA', 'BBAS3.SA', 'ITUB4.SA', 'NORD3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['EQPA6.SA', 'OIBR4.SA', 'SLED3.SA', 'CLSC4.SA', 'GPIV33.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['CESP3.SA', 'SANB11.SA', 'GPCP3.SA', 'BEES4.SA', 'TRIS3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['CRIV3.SA', 'LAME3.SA', 'ENBR3.SA', 'INEP4.SA', 'BRDT3.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['PINE4.SA', 'EMAE4.SA', 'SEER3.SA', 'PTBL3.SA', 'JBDU4.SA']\n",
      "Calculando semana 2019-01-01 - 2019-12-31\n",
      "Tickers = ['JFEN3.SA', 'OMGE3.SA', 'CPLE3.SA', 'CCPR3.SA', 'ALUP3.SA']\n"
     ]
    }
   ],
   "source": [
    "# chamada da função monkey_stocks()\n",
    "ms = monkey_stocks(data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Visualização\n",
    "\n",
    "A seguir um gráfico do retorno acumulado da estratégia no tempo."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Retorno Acumulado')"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x648 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(16,9))\n",
    "ax = (1 + ms['mean']).cumprod().plot(title='Retorno acumulado do Monkey Stocks')\n",
    "ax.set_xlabel(\"Tempo\")\n",
    "ax.set_ylabel(\"Retorno Acumulado\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": ".quantenv",
   "language": "python",
   "name": ".quantenv"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}