In [50]:

    

# code written in py_3.0

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import matplotlib.dates as mdates
import seaborn as sns


    

In [44]:

    

# find path to your SwordForecasting.xlsx
df_sales = pd.read_excel(open('C:/Users/craigrshenton/Desktop/Dropbox/excel_data_sci/ch08/SwordForecasting.xlsm','rb'), sheetname=0)
df_sales = df_sales.iloc[0:36, 0:2]
df_sales.rename(columns={'t':'Time'}, inplace=True)
df_sales.head()


    

    
Out[44]:






  

    

      

      
Time
      
Demand
    
  
  

    

      
0
      
1
      
165.0
    
    

      
1
      
2
      
171.0
    
    

      
2
      
3
      
147.0
    
    

      
3
      
4
      
143.0
    
    

      
4
      
5
      
164.0
    
  

In [45]:

    

df_sales.Time = pd.date_range('2010-1', periods=len(df_sales.Time), freq='M') # 'Time' is now in time-series format
df_sales = df_sales.set_index('Time') # set Time as Series index


    

In [76]:

    

sns.set(style="darkgrid", context="notebook", font_scale=0.9, rc={"lines.linewidth": 1.5}) # make plots look nice

fig, ax = plt.subplots(1)
ax.plot(df_sales)
plt.ylabel('Demand')
plt.xlabel('Date')

# rotate and align the tick labels so they look better
fig.autofmt_xdate()

# use a more precise date string for the x axis locations
ax.fmt_xdata = mdates.DateFormatter('%Y-%m-%d')

plt.show()


    

In [77]:

    

from statsmodels.tsa.stattools import adfuller
def test_stationarity(timeSeries):

    fig, ax = plt.subplots(1)
    ax.plot(timeSeries, '-.', label='raw data')
    ax.plot(timeSeries.rolling(12).mean(), label='moving average (year)')
    ax.plot(timeSeries.expanding().mean(), label='expanding')
    ax.plot(timeSeries.ewm(alpha=0.03).mean(), label='EWMA ($\\alpha=.03$)')

    # rotate and align the tick labels so they look better
    fig.autofmt_xdate()
    
    plt.ylabel('Demand')
    plt.xlabel('Date')
    plt.legend(bbox_to_anchor=(1.35, .5))
    plt.show()
    
    # perform Dickey-Fuller test:
    print('Results of Dickey-Fuller Test:')
    dftest = adfuller(timeSeries.ix[:,0], autolag='AIC')
    dfoutput = pd.Series(dftest[0:4], index=['Test Statistic','p-value','#Lags Used','Number of Observations Used'])
    for key,value in dftest[4].items():
        dfoutput['Critical Value (%s)'%key] = value
    print(dfoutput)
    
test_stationarity(df_sales)


    

    













    




Results of Dickey-Fuller Test:
Test Statistic                  2.189225
p-value                         0.998874
#Lags Used                     10.000000
Number of Observations Used    25.000000
Critical Value (1%)            -3.723863
Critical Value (10%)           -2.632800
Critical Value (5%)            -2.986489
dtype: float64

In [ ]:

    




    

In [ ]:

    




    

In [161]:

    

def tsplot(y, lags=None, figsize=(10, 8)):
    fig = plt.figure(figsize=figsize)
    layout = (2, 2)
    ts_ax = plt.subplot2grid(layout, (0, 0), colspan=2)
    acf_ax = plt.subplot2grid(layout, (1, 0))
    pacf_ax = plt.subplot2grid(layout, (1, 1))

    y.plot(ax=ts_ax)
    smt.graphics.plot_acf(y, lags=lags, ax=acf_ax)
    smt.graphics.plot_pacf(y, lags=lags, ax=pacf_ax)
    [ax.set_xlim(1.5) for ax in [acf_ax, pacf_ax]]
    sns.despine()
    plt.tight_layout()
    return ts_ax, acf_ax, pacf_ax


    

In [170]:

    

import statsmodels.formula.api as smf
import statsmodels.tsa.api as smt
import statsmodels.api as sm

mod = smt.ARIMA(df_sales, order=(1, 1, 1))
res = mod.fit()


    

In [181]:

    

pred_dy = res.get_prediction(start=min(df_sales.index), dynamic=min(df_sales.index))
pred_dy_ci = pred_dy.conf_int()


    

    




---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-181-22ecc5b05bf7> in <module>()
----> 1 pred_dy = res.get_prediction(start=min(df_sales.index), dynamic=min(df_sales.index))
      2 pred_dy_ci = pred_dy.conf_int()

D:\Anaconda\lib\site-packages\statsmodels\base\wrapper.py in __getattribute__(self, attr)
     33             pass
     34 
---> 35         obj = getattr(results, attr)
     36         data = results.model.data
     37         how = self._wrap_attrs.get(attr)

AttributeError: 'ARIMAResults' object has no attribute 'get_prediction'

In [ ]: