.. _pyensaeStockPricesrst:

=============================================================
Manipulation de séries financières avec la classe StockPrices
=============================================================


.. only:: html

    **Links:** :download:`notebook <pyensae_StockPrices.ipynb>`, :downloadlink:`html <pyensae_StockPrices2html.html>`, :download:`PDF <pyensae_StockPrices.pdf>`, :download:`python <pyensae_StockPrices.py>`, :downloadlink:`slides <pyensae_StockPrices.slides.html>`, :githublink:`GitHub|_doc/notebooks/pyensae_StockPrices.ipynb|*`


La classe
`StockPrices <http://www.xavierdupre.fr/app/pyensae/helpsphinx/notebooks/pyensae%20StockPrices.html?highlight=stockprices>`__
facilite la récupération de données financières via différents sites. Le
site `Yahoo Finance <http://finance.yahoo.com/>`__ requiet maintenant un
cookie (depuis Mai 2017) et il est préférable de choisir
`Google <https://www.google.com/finance?q=apple&ei=bPghWZnLDoKVULfBk-gC>`__
ou `Quandl <https://www.quandl.com/tools/python>`__. Google ne
fonctionne que les marchés américains,
`Quandl <https://www.quandl.com/tools/python>`__ a des historiques plus
courts.

.. code:: ipython3

    import pyensae
    from jyquickhelper import add_notebook_menu
    add_notebook_menu()






.. contents::
    :local:





.. code:: ipython3

    %matplotlib inline
    import matplotlib.pyplot as plt
    plt.style.use('ggplot')

Initialisation
~~~~~~~~~~~~~~

.. code:: ipython3

    import pyensae
    import os
    from pyensae.finance import StockPrices
    cache = os.path.abspath("cache")
    if not os.path.exists(cache):
        os.mkdir(cache)

Créer un objet StockPrices
~~~~~~~~~~~~~~~~~~~~~~~~~~

Le plus est d’utiliser le tick de la série financière utilisé par le
site `Yahoo Finance <https://fr.finance.yahoo.com>`__ ou `Google
Finance <https://www.google.com/search?q=finance>`__ qui fait maintenant
partie du moteur de recherche ou
`quandl <https://www.quandl.com/tools/python>`__.

.. code:: ipython3

    source = 'yahoo_new'
    tick = 'MSFT'
    stock = StockPrices(tick, folder=cache, url=source)
    stock.head()






.. raw:: html

    <div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }

        .dataframe tbody tr th {
            vertical-align: top;
        }

        .dataframe thead th {
            text-align: right;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr style="text-align: right;">
          <th></th>
          <th>Date</th>
          <th>Open</th>
          <th>High</th>
          <th>Low</th>
          <th>Close</th>
          <th>Adj Close</th>
          <th>Volume</th>
        </tr>
        <tr>
          <th>Date</th>
          <th></th>
          <th></th>
          <th></th>
          <th></th>
          <th></th>
          <th></th>
          <th></th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>2000-01-03</th>
          <td>2000-01-03</td>
          <td>58.68750</td>
          <td>59.3125</td>
          <td>56.00000</td>
          <td>58.28125</td>
          <td>42.295185</td>
          <td>53228400</td>
        </tr>
        <tr>
          <th>2000-01-04</th>
          <td>2000-01-04</td>
          <td>56.78125</td>
          <td>58.5625</td>
          <td>56.12500</td>
          <td>56.31250</td>
          <td>40.866425</td>
          <td>54119000</td>
        </tr>
        <tr>
          <th>2000-01-05</th>
          <td>2000-01-05</td>
          <td>55.56250</td>
          <td>58.1875</td>
          <td>54.68750</td>
          <td>56.90625</td>
          <td>41.297348</td>
          <td>64059600</td>
        </tr>
        <tr>
          <th>2000-01-06</th>
          <td>2000-01-06</td>
          <td>56.09375</td>
          <td>56.9375</td>
          <td>54.18750</td>
          <td>55.00000</td>
          <td>39.913952</td>
          <td>54976600</td>
        </tr>
        <tr>
          <th>2000-01-07</th>
          <td>2000-01-07</td>
          <td>54.31250</td>
          <td>56.1250</td>
          <td>53.65625</td>
          <td>55.71875</td>
          <td>40.435547</td>
          <td>62013600</td>
        </tr>
      </tbody>
    </table>
    </div>



.. code:: ipython3

    stock.tail()






.. raw:: html

    <div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }

        .dataframe tbody tr th {
            vertical-align: top;
        }

        .dataframe thead th {
            text-align: right;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr style="text-align: right;">
          <th></th>
          <th>Date</th>
          <th>Open</th>
          <th>High</th>
          <th>Low</th>
          <th>Close</th>
          <th>Adj Close</th>
          <th>Volume</th>
        </tr>
        <tr>
          <th>Date</th>
          <th></th>
          <th></th>
          <th></th>
          <th></th>
          <th></th>
          <th></th>
          <th></th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>2019-01-25</th>
          <td>2019-01-25</td>
          <td>107.239998</td>
          <td>107.879997</td>
          <td>106.199997</td>
          <td>107.169998</td>
          <td>107.169998</td>
          <td>31225600</td>
        </tr>
        <tr>
          <th>2019-01-28</th>
          <td>2019-01-28</td>
          <td>106.260002</td>
          <td>106.480003</td>
          <td>104.660004</td>
          <td>105.080002</td>
          <td>105.080002</td>
          <td>29476700</td>
        </tr>
        <tr>
          <th>2019-01-29</th>
          <td>2019-01-29</td>
          <td>104.879997</td>
          <td>104.970001</td>
          <td>102.169998</td>
          <td>102.940002</td>
          <td>102.940002</td>
          <td>31490500</td>
        </tr>
        <tr>
          <th>2019-01-30</th>
          <td>2019-01-30</td>
          <td>104.620003</td>
          <td>106.379997</td>
          <td>104.330002</td>
          <td>106.379997</td>
          <td>106.379997</td>
          <td>49471900</td>
        </tr>
        <tr>
          <th>2019-01-31</th>
          <td>2019-01-31</td>
          <td>103.800003</td>
          <td>105.220001</td>
          <td>103.180000</td>
          <td>104.430000</td>
          <td>104.430000</td>
          <td>55636400</td>
        </tr>
      </tbody>
    </table>
    </div>



La classe StockPrices contient un objet pandas.DataFrame auquel on
accède en écrivant stock.dataframe ou stock.df :

.. code:: ipython3

    stock.dataframe.columns




.. parsed-literal::
    Index(['Date', 'Open', 'High', 'Low', 'Close', 'Adj Close', 'Volume'], dtype='object')



De la même manière, on peut créer un objet StockPrices à partir d’un
DataFrame :

.. code:: ipython3

    import pandas
    data = [{"Date":"2014-04-01", "Close":105.6}, {"Date":"2014-04-02", "Close":104.6}, 
            {"Date":"2014-04-03", "Close":105.8}, ]
    df = pandas.DataFrame(data)
    stock = StockPrices("donnees",df)
    stock.head()






.. raw:: html

    <div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }

        .dataframe tbody tr th {
            vertical-align: top;
        }

        .dataframe thead th {
            text-align: right;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr style="text-align: right;">
          <th></th>
          <th>Close</th>
          <th>Date</th>
        </tr>
        <tr>
          <th>Date</th>
          <th></th>
          <th></th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>2014-04-01</th>
          <td>105.6</td>
          <td>2014-04-01</td>
        </tr>
        <tr>
          <th>2014-04-02</th>
          <td>104.6</td>
          <td>2014-04-02</td>
        </tr>
        <tr>
          <th>2014-04-03</th>
          <td>105.8</td>
          <td>2014-04-03</td>
        </tr>
      </tbody>
    </table>
    </div>



Quelques graphes
~~~~~~~~~~~~~~~~

Premier dessin, on télécharge les données de BNP puis on dessine le
cours de l’action.

.. code:: ipython3

    import datetime
    stock = StockPrices(tick, folder=cache, url=source)
    ax = StockPrices.draw(stock, figsize=(12,6))



.. image:: pyensae_StockPrices_15_0.png


.. code:: ipython3

    stock = StockPrices(tick, folder=cache, url=source)
    StockPrices.draw(stock, figsize=(12,6));



.. image:: pyensae_StockPrices_16_0.png


La même chose se produit sur une autre série financière mais pas à la
même date. On trace maintenant la série *Open* (*Adj Close* défini sur
cette page `View and download historical price, dividend, or split
data <https://help.yahoo.com/kb/finance/historical-prices-sln2311.html?impressions=true>`__
n’est disponible qu’avec Yahoo).

.. code:: ipython3

    stock = StockPrices("MSFT", folder=cache, url='yahoo')
    StockPrices.draw(stock, field=["Open", "Close"], figsize=(12,6));



.. image:: pyensae_StockPrices_18_0.png


Ce type de série ne fait pas toujours apparaître les saut de prix qui
survient comme par exemple le 20 février 2002 lorsque le cours nominal
de l’action de la BNP a été divisé par deux pour augmenter la liquidité.
Le nombre d’actions a été multiplié par deux. Les données sont le plus
souvent corrigées `BNP février
2002 <http://www.google.com/finance/historical?cid=667923&startdate=Feb+15%2C+2002&enddate=Feb+28%2C+2002&num=30&ei=AP8hWaDXF4WNUNuujYgK>`__.

Ajouter une seconde série sur un graphe
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Dans l’exemple suivant, on trace une série financière puis on ajoute la
série des rendements sur un second axe.

.. code:: ipython3

    stock.head()






.. raw:: html

    <div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }

        .dataframe tbody tr th {
            vertical-align: top;
        }

        .dataframe thead th {
            text-align: right;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr style="text-align: right;">
          <th></th>
          <th>Date</th>
          <th>Open</th>
          <th>High</th>
          <th>Low</th>
          <th>Close</th>
          <th>Adj Close</th>
          <th>Volume</th>
        </tr>
        <tr>
          <th>Date</th>
          <th></th>
          <th></th>
          <th></th>
          <th></th>
          <th></th>
          <th></th>
          <th></th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>2000-01-03</th>
          <td>2000-01-03</td>
          <td>58.68750</td>
          <td>59.3125</td>
          <td>56.00000</td>
          <td>58.28125</td>
          <td>42.295185</td>
          <td>53228400</td>
        </tr>
        <tr>
          <th>2000-01-04</th>
          <td>2000-01-04</td>
          <td>56.78125</td>
          <td>58.5625</td>
          <td>56.12500</td>
          <td>56.31250</td>
          <td>40.866425</td>
          <td>54119000</td>
        </tr>
        <tr>
          <th>2000-01-05</th>
          <td>2000-01-05</td>
          <td>55.56250</td>
          <td>58.1875</td>
          <td>54.68750</td>
          <td>56.90625</td>
          <td>41.297348</td>
          <td>64059600</td>
        </tr>
        <tr>
          <th>2000-01-06</th>
          <td>2000-01-06</td>
          <td>56.09375</td>
          <td>56.9375</td>
          <td>54.18750</td>
          <td>55.00000</td>
          <td>39.913952</td>
          <td>54976600</td>
        </tr>
        <tr>
          <th>2000-01-07</th>
          <td>2000-01-07</td>
          <td>54.31250</td>
          <td>56.1250</td>
          <td>53.65625</td>
          <td>55.71875</td>
          <td>40.435547</td>
          <td>62013600</td>
        </tr>
      </tbody>
    </table>
    </div>



.. code:: ipython3

    stock = StockPrices(tick)
    ret = stock.returns()["2019-01-04":"2019-02-02"]
    ret.dataframe.loc["2019-01-11":"2019-01-18","Close"] = 0  # on annule certains valeurs
    ax = stock.plot(figsize=(16, 5))
    ret.plot(axis=2, ax=ax, label_prefix="r", color='blue');



.. image:: pyensae_StockPrices_23_0.png


Quelques opérations
~~~~~~~~~~~~~~~~~~~

.. code:: ipython3

    os.listdir(cache)




.. parsed-literal::
    ['GOOGL.2000-01-03.2018-03-15.txt',
     'GOOGL.2000-01-03.2018-05-13.txt',
     'GOOGL.2000-01-03.2019-02-01.txt',
     'MSFT.2000-01-03.2018-03-15.txt',
     'MSFT.2000-01-03.2018-05-13.txt',
     'MSFT.2000-01-03.2019-02-01.txt']



On affiche les dernières lignes.

.. code:: ipython3

    stock.tail()






.. raw:: html

    <div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }

        .dataframe tbody tr th {
            vertical-align: top;
        }

        .dataframe thead th {
            text-align: right;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr style="text-align: right;">
          <th></th>
          <th>Date</th>
          <th>Open</th>
          <th>High</th>
          <th>Low</th>
          <th>Close</th>
          <th>Adj Close</th>
          <th>Volume</th>
        </tr>
        <tr>
          <th>Date</th>
          <th></th>
          <th></th>
          <th></th>
          <th></th>
          <th></th>
          <th></th>
          <th></th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>2019-01-25</th>
          <td>2019-01-25</td>
          <td>107.239998</td>
          <td>107.879997</td>
          <td>106.199997</td>
          <td>107.169998</td>
          <td>107.169998</td>
          <td>31225600</td>
        </tr>
        <tr>
          <th>2019-01-28</th>
          <td>2019-01-28</td>
          <td>106.260002</td>
          <td>106.480003</td>
          <td>104.660004</td>
          <td>105.080002</td>
          <td>105.080002</td>
          <td>29476700</td>
        </tr>
        <tr>
          <th>2019-01-29</th>
          <td>2019-01-29</td>
          <td>104.879997</td>
          <td>104.970001</td>
          <td>102.169998</td>
          <td>102.940002</td>
          <td>102.940002</td>
          <td>31490500</td>
        </tr>
        <tr>
          <th>2019-01-30</th>
          <td>2019-01-30</td>
          <td>104.620003</td>
          <td>106.379997</td>
          <td>104.330002</td>
          <td>106.379997</td>
          <td>106.379997</td>
          <td>49471900</td>
        </tr>
        <tr>
          <th>2019-01-31</th>
          <td>2019-01-31</td>
          <td>103.800003</td>
          <td>105.220001</td>
          <td>103.180000</td>
          <td>104.430000</td>
          <td>104.430000</td>
          <td>55636400</td>
        </tr>
      </tbody>
    </table>
    </div>



On récupère la série des rendements.

.. code:: ipython3

    ret = stock.returns()
    ret.tail()






.. raw:: html

    <div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }

        .dataframe tbody tr th {
            vertical-align: top;
        }

        .dataframe thead th {
            text-align: right;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr style="text-align: right;">
          <th></th>
          <th>Date</th>
          <th>Volume</th>
          <th>Open</th>
          <th>High</th>
          <th>Low</th>
          <th>Close</th>
          <th>Adj Close</th>
        </tr>
        <tr>
          <th>Date</th>
          <th></th>
          <th></th>
          <th></th>
          <th></th>
          <th></th>
          <th></th>
          <th></th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>2019-01-25</th>
          <td>2019-01-25</td>
          <td>31225600</td>
          <td>0.003556</td>
          <td>0.008224</td>
          <td>0.008164</td>
          <td>0.009134</td>
          <td>0.009134</td>
        </tr>
        <tr>
          <th>2019-01-28</th>
          <td>2019-01-28</td>
          <td>29476700</td>
          <td>-0.009138</td>
          <td>-0.012977</td>
          <td>-0.014501</td>
          <td>-0.019502</td>
          <td>-0.019502</td>
        </tr>
        <tr>
          <th>2019-01-29</th>
          <td>2019-01-29</td>
          <td>31490500</td>
          <td>-0.012987</td>
          <td>-0.014181</td>
          <td>-0.023791</td>
          <td>-0.020365</td>
          <td>-0.020365</td>
        </tr>
        <tr>
          <th>2019-01-30</th>
          <td>2019-01-30</td>
          <td>49471900</td>
          <td>-0.002479</td>
          <td>0.013432</td>
          <td>0.021141</td>
          <td>0.033417</td>
          <td>0.033417</td>
        </tr>
        <tr>
          <th>2019-01-31</th>
          <td>2019-01-31</td>
          <td>55636400</td>
          <td>-0.007838</td>
          <td>-0.010904</td>
          <td>-0.011023</td>
          <td>-0.018330</td>
          <td>-0.018330</td>
        </tr>
      </tbody>
    </table>
    </div>



On trace la série des rendements pour les derniers mois.

.. code:: ipython3

    StockPrices.draw(ret, figsize=(12,6), begin="2013-12-01", date_format="%Y-%m");



.. image:: pyensae_StockPrices_31_0.png


Quelques notions sur les dates
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

La classe StockPrices utilise les dates sous forme de chaînes de
caractères. De cette façon, il n’est pas possible de faire des
opérations dessus. Pour ce faire, il faut les convertir en un objet
appelé datetime.

.. code:: ipython3

    from datetime import datetime, timedelta
    dt = datetime.strptime("2014-03-31","%Y-%m-%d")
    dt




.. parsed-literal::
    datetime.datetime(2014, 3, 31, 0, 0)



On ajoute un jour :

.. code:: ipython3

    delta = timedelta(1)
    dt = dt + delta
    dt




.. parsed-literal::
    datetime.datetime(2014, 4, 1, 0, 0)



Puis on convertit dans l’autre sens :

.. code:: ipython3

    s = dt.strftime("%Y-%m-%d")
    s




.. parsed-literal::
    '2014-04-01'



Promenade dans l’index
~~~~~~~~~~~~~~~~~~~~~~

Il est facile de récupérer les valeurs correspondant à une date précise.
Mais comment récupérer la valeur du jour d’après ?

.. code:: ipython3

    tick2 = 'GOOGL'
    stock = StockPrices(tick2, folder=cache, url=source)
    df = stock.dataframe
    print("A", df["2005-01-04":"2005-01-06"])
    print("D", df.loc["2005-01-04","Close"])
    print("G", df.index.get_loc("2005-01-06"))  # retourne la position de cette date


.. parsed-literal::
    A                   Date        Open        High        Low      Close  \
    Date                                                                   
    2005-01-04  2005-01-04  100.800804  101.566566  96.836838  97.347351   
    2005-01-05  2005-01-05   96.821823   98.548546  96.211212  96.851852   
    2005-01-06  2005-01-06   97.637634   98.048050  93.953957  94.369370   
                Adj Close    Volume  
    Date                             
    2005-01-04  97.347351  27484200  
    2005-01-05  96.851852  16456700  
    2005-01-06  94.369370  20753400  
    D 97.347351
    G 97


Sauver les tables
~~~~~~~~~~~~~~~~~

On peut conserver les données sous forme de fichiers pour les récupérer
plus tard.

.. code:: ipython3

    stock = StockPrices(tick2, folder=cache, url=source)
    stock.dataframe.to_csv("donnees.txt", sep="\t")
    [_ for _ in os.listdir(".") if "donnees" in _]




.. parsed-literal::
    ['donnees.txt']



Le fichier est sauvé. Pour le récupérer avec pandas :

.. code:: ipython3

    import pandas
    df = pandas.read_csv("donnees.txt", sep="\t")
    df.head()






.. raw:: html

    <div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }

        .dataframe tbody tr th {
            vertical-align: top;
        }

        .dataframe thead th {
            text-align: right;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr style="text-align: right;">
          <th></th>
          <th>Date</th>
          <th>Date.1</th>
          <th>Open</th>
          <th>High</th>
          <th>Low</th>
          <th>Close</th>
          <th>Adj Close</th>
          <th>Volume</th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>0</th>
          <td>2004-08-19</td>
          <td>2004-08-19</td>
          <td>50.050049</td>
          <td>52.082081</td>
          <td>48.028027</td>
          <td>50.220219</td>
          <td>50.220219</td>
          <td>44659000</td>
        </tr>
        <tr>
          <th>1</th>
          <td>2004-08-20</td>
          <td>2004-08-20</td>
          <td>50.555557</td>
          <td>54.594593</td>
          <td>50.300301</td>
          <td>54.209209</td>
          <td>54.209209</td>
          <td>22834300</td>
        </tr>
        <tr>
          <th>2</th>
          <td>2004-08-23</td>
          <td>2004-08-23</td>
          <td>55.430431</td>
          <td>56.796795</td>
          <td>54.579578</td>
          <td>54.754753</td>
          <td>54.754753</td>
          <td>18256100</td>
        </tr>
        <tr>
          <th>3</th>
          <td>2004-08-24</td>
          <td>2004-08-24</td>
          <td>55.675674</td>
          <td>55.855854</td>
          <td>51.836838</td>
          <td>52.487488</td>
          <td>52.487488</td>
          <td>15247300</td>
        </tr>
        <tr>
          <th>4</th>
          <td>2004-08-25</td>
          <td>2004-08-25</td>
          <td>52.532532</td>
          <td>54.054054</td>
          <td>51.991993</td>
          <td>53.053055</td>
          <td>53.053055</td>
          <td>9188600</td>
        </tr>
      </tbody>
    </table>
    </div>



Les dates apparaissent deux fois.

.. code:: ipython3

    with open("donnees.txt","r") as f: 
        text = f.read()
    print(text[:400])


.. parsed-literal::
    Date	Date	Open	High	Low	Close	Adj Close	Volume
    2004-08-19	2004-08-19	50.050049	52.082081	48.028027	50.220219	50.220219	44659000
    2004-08-20	2004-08-20	50.555557	54.594593	50.300301	54.209209	54.209209	22834300
    2004-08-23	2004-08-23	55.43043100000001	56.796795	54.57957800000001	54.754753	54.754753	18256100
    2004-08-24	2004-08-24	55.675674	55.855854	51.836838	52.487488	52.487488	15247300
    2004-08-25	20


Cela est dû au fait que les dates sont à la fois une colonne et servent
d’index. Pour éviter de les conserver deux fois, on demande
explicitement à ce que l’index ne soit pas ajouté au fichier :

.. code:: ipython3

    stock = StockPrices(tick2, folder=cache, url=source)
    stock.dataframe.to_csv("donnees.txt", sep="\t", index=False)

Puis on récupère les données :

.. code:: ipython3

    df = pandas.read_csv("donnees.txt",sep="\t")
    df.head()






.. raw:: html

    <div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }

        .dataframe tbody tr th {
            vertical-align: top;
        }

        .dataframe thead th {
            text-align: right;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr style="text-align: right;">
          <th></th>
          <th>Date</th>
          <th>Open</th>
          <th>High</th>
          <th>Low</th>
          <th>Close</th>
          <th>Adj Close</th>
          <th>Volume</th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>0</th>
          <td>2004-08-19</td>
          <td>50.050049</td>
          <td>52.082081</td>
          <td>48.028027</td>
          <td>50.220219</td>
          <td>50.220219</td>
          <td>44659000</td>
        </tr>
        <tr>
          <th>1</th>
          <td>2004-08-20</td>
          <td>50.555557</td>
          <td>54.594593</td>
          <td>50.300301</td>
          <td>54.209209</td>
          <td>54.209209</td>
          <td>22834300</td>
        </tr>
        <tr>
          <th>2</th>
          <td>2004-08-23</td>
          <td>55.430431</td>
          <td>56.796795</td>
          <td>54.579578</td>
          <td>54.754753</td>
          <td>54.754753</td>
          <td>18256100</td>
        </tr>
        <tr>
          <th>3</th>
          <td>2004-08-24</td>
          <td>55.675674</td>
          <td>55.855854</td>
          <td>51.836838</td>
          <td>52.487488</td>
          <td>52.487488</td>
          <td>15247300</td>
        </tr>
        <tr>
          <th>4</th>
          <td>2004-08-25</td>
          <td>52.532532</td>
          <td>54.054054</td>
          <td>51.991993</td>
          <td>53.053055</td>
          <td>53.053055</td>
          <td>9188600</td>
        </tr>
      </tbody>
    </table>
    </div>



On vérifie le fichier sur disque dur :

.. code:: ipython3

    with open("donnees.txt", "r") as f:
        text = f.read()
    print(text[:400])


.. parsed-literal::
    Date	Open	High	Low	Close	Adj Close	Volume
    2004-08-19	50.050049	52.082081	48.028027	50.220219	50.220219	44659000
    2004-08-20	50.555557	54.594593	50.300301	54.209209	54.209209	22834300
    2004-08-23	55.43043100000001	56.796795	54.57957800000001	54.754753	54.754753	18256100
    2004-08-24	55.675674	55.855854	51.836838	52.487488	52.487488	15247300
    2004-08-25	52.532532	54.054054	51.991993	53.05305500000001	53.


C’est mieux.