Scribd Logo
Upload

Welcome to Scribd!

  • 33 views

    Pronóstico: Leer Datos

    Uploaded by

    Ricardo Sebastian
    The document discusses using an ARIMA model to forecast time series data. It loads data on past values, splits the data into training and test sets, fits an auto_arima model to the training data, and uses the model to make a 10-day forecast of future values. It then plots the forecast values against the actual future values to compare the predictive performance of the ARIMA model.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Pronóstico: Leer Datos

    Uploaded by

    Ricardo Sebastian
    33 views5 pages
    The document discusses using an ARIMA model to forecast time series data. It loads data on past values, splits the data into training and test sets, fits an auto_arima model to the training data, and uses the model to make a 10-day forecast of future values. It then plots the forecast values against the actual future values to compare the predictive performance of the ARIMA model.

    Original Title

    Pronostico

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    The document discusses using an ARIMA model to forecast time series data. It loads data on past values, splits the data into training and test sets, fits an auto_arima model to the training data, and uses the model to make a 10-day forecast of future values. It then plots the forecast values against the actual future values to compare the predictive performance of the ARIMA model.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    33 views5 pages

    Pronóstico: Leer Datos

    Uploaded by

    Ricardo Sebastian
    The document discusses using an ARIMA model to forecast time series data. It loads data on past values, splits the data into training and test sets, fits an auto_arima model to the training data, and uses the model to make a 10-day forecast of future values. It then plots the forecast values against the actual future values to compare the predictive performance of the ARIMA model.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    You are on page 1of 5

    1/8/2021 Untitled1.

    ipynb - Colaboratory

    Pronóstico

    import pandas as pd
    import numpy as np
    import seaborn as sns
    from matplotlib import pyplot as plt
    from fbprophet import Prophet

    Leer Datos

    df=pd.read_csv("./Datos4.csv")
    df.head()
     

    Fecha Datos

    0 01-01-2021 6

    1 02-01-2021 31

    2 03-01-2021 4

    3 04-01-2021 143

    4 05-01-2021 137

    Crear Datos de Test y de Entrenamiento

    p_train=0.8
     
    train = df[:int((len(df))*p_train)] 
    test = df[int((len(df))*p_train):]

    df_comp=train.copy()

    df_comp.head()

    https://colab.research.google.com/drive/1RH1BvQGwhJY1qzwwKWqhaVDfSpMsxkhb#scrollTo=ns2nesm5n9mS&printMode=true 1/5
    1/8/2021 Untitled1.ipynb - Colaboratory

    Fecha Datos

    0 01-01-2021 6

    1 02-01-2021 31

    2 03-01-2021 4

    3 04-01-2021 143

    4 05-01-2021 137
    df_comp.Fecha=pd.to_datetime(df_comp.Fecha, dayfirst= True)
    df_comp.set_index("Fecha",inplace=True)

    df_comp.head()

    Datos

    Fecha

    2021-01-01 6

    2021-01-02 31

    2021-01-03 4

    2021-01-04 143

    2021-01-05 137

    df.Datos.plot(figsize=(10,5))
    plt.title("Datos", size=24)
    plt.show()

      !pip install pmdarima
    https://colab.research.google.com/drive/1RH1BvQGwhJY1qzwwKWqhaVDfSpMsxkhb#scrollTo=ns2nesm5n9mS&printMode=true 2/5
    1/8/2021 Untitled1.ipynb - Colaboratory
    p p p
      from pmdarima.arima import auto_arima

    Collecting pmdarima

    Downloading pmdarima-1.8.2-cp37-cp37m-manylinux1_x86_64.whl (1.5 MB)

    |████████████████████████████████| 1.5 MB 4.8 MB/s

    Requirement already satisfied: joblib>=0.11 in /usr/local/lib/python3.7/dist-packages (


    Requirement already satisfied: numpy~=1.19.0 in /usr/local/lib/python3.7/dist-packages
    Requirement already satisfied: setuptools!=50.0.0,>=38.6.0 in /usr/local/lib/python3.7/
    Requirement already satisfied: scipy>=1.3.2 in /usr/local/lib/python3.7/dist-packages (
    Requirement already satisfied: urllib3 in /usr/local/lib/python3.7/dist-packages (from
    Requirement already satisfied: Cython!=0.29.18,>=0.29 in /usr/local/lib/python3.7/dist-
    Requirement already satisfied: scikit-learn>=0.22 in /usr/local/lib/python3.7/dist-pack
    Requirement already satisfied: pandas>=0.19 in /usr/local/lib/python3.7/dist-packages (
    Collecting statsmodels!=0.12.0,>=0.11

    Downloading statsmodels-0.12.2-cp37-cp37m-manylinux1_x86_64.whl (9.5 MB)

    |████████████████████████████████| 9.5 MB 19.9 MB/s

    Requirement already satisfied: python-dateutil>=2.7.3 in /usr/local/lib/python3.7/dist-


    Requirement already satisfied: pytz>=2017.2 in /usr/local/lib/python3.7/dist-packages (
    Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.7/dist-packages (from
    Requirement already satisfied: patsy>=0.5 in /usr/local/lib/python3.7/dist-packages (fr
    Installing collected packages: statsmodels, pmdarima

    Attempting uninstall: statsmodels

    Found existing installation: statsmodels 0.10.2

    Uninstalling statsmodels-0.10.2:

    Successfully uninstalled statsmodels-0.10.2

    Successfully installed pmdarima-1.8.2 statsmodels-0.12.2

    WARNING: The following packages were previously imported in this runtime:

    [statsmodels]

    You must restart the runtime in order to use newly installed versions.

    RESTART RUNTIME

    model_auto=auto_arima(df_comp.Datos, m=5,
                          max_order= None, max_p=7, max_q=7, max_d=2, max_P=4, max_Q=4, max_D=2, 
                          maxiter=50, alpha=0.05, information_criterion='aic')

    model_auto

    ARIMA(maxiter=50, method='lbfgs', order=(0, 0, 1), out_of_sample_size=0,

    scoring='mse', scoring_args={}, seasonal_order=(0, 0, 1, 5),

    start_params=None, suppress_warnings=True, trend=None,

    with_intercept=True)

    model_auto.summary()

    https://colab.research.google.com/drive/1RH1BvQGwhJY1qzwwKWqhaVDfSpMsxkhb#scrollTo=ns2nesm5n9mS&printMode=true 3/5
    1/8/2021 Untitled1.ipynb - Colaboratory

    SARIMAX Results
    Dep. Variable: y No. Observations: 167
    Model: SARIMAX(0, 0, 1)x(0, 0, 1, 5) Log Likelihood -841.427
    Date: Sun, 01 Aug 2021 AIC 1690.855
    Time: 16:25:43 BIC 1703.327
    Sample: 0 HQIC 1695.917
    - 167
    Covariance Type: opg
    coef std err z P>|z| [0.025 0.975]
    intercept 74.5716 1.797 41.499 0.000 71.050 78.094
    ma.L1 0.2953 0.094 3.126 0.002 0.110 0.480
    ma.S.L5 -0.5747 0.074 -7.756 0.000 -0.720 -0.429
    sigma2 1375.4178 175.577 7.834 0.000 1031.293 1719.543
    Ljung-Box (L1) (Q): 0.02 Jarque-Bera (JB): 5.70
    Prob(Q): 0.89 Prob(JB): 0.06
    Heteroskedasticity (H): 1.02
    date=pd.to_datetime("06-17-2021") Skew: 0.44
    date Prob(H) (two-sided): 0.93 Kurtosis: 3.21

    Timestamp('2021-06-17 00:00:00')
    Warnings:

    [1] Covariance matrix calculated using the outer product of gradients (complex-step).
    i_date=date+pd.to_timedelta(np.arange(10),'D')
    i_date

    DatetimeIndex(['2021-06-17', '2021-06-18', '2021-06-19', '2021-06-20',

    '2021-06-21', '2021-06-22', '2021-06-23', '2021-06-24',

    '2021-06-25', '2021-06-26'],

    dtype='datetime64[ns]', freq=None)

    pronostico=model_auto.predict(len(i_date))

    df_test=pd.Series(pronostico,index=i_date)
    df_test.head()

    2021-06-17 93.366008

    2021-06-18 114.026628

    2021-06-19 47.533056

    2021-06-20 55.214624

    2021-06-21 66.332862

    dtype: float64

    df_auto_pred=pd.DataFrame({'Pronosticos': pronostico},index=i_date)

    df_auto_pred.plot(figsize=(10,5),color="red")
    df_comp.Datos.plot(color="blue")
    plt.title("Pronostico vs Real", size=24)
    plt.show()

    https://colab.research.google.com/drive/1RH1BvQGwhJY1qzwwKWqhaVDfSpMsxkhb#scrollTo=ns2nesm5n9mS&printMode=true 4/5
    1/8/2021 Untitled1.ipynb - Colaboratory

    https://colab.research.google.com/drive/1RH1BvQGwhJY1qzwwKWqhaVDfSpMsxkhb#scrollTo=ns2nesm5n9mS&printMode=true 5/5

    You might also like