In [1]:

import pandas as pd
import numpy as np
from dateutil.parser import parse
import datetime
from pandas_datareader import data, wb

In [2]:

df = pd.read_csv('beatles.csv')

In [3]:

df.dtypes

Out[3]:

Unnamed: 0              int64
Title                  object
Released               object
Label                  object
UK Chart Position     float64
US Chart Position     float64
BPI Certification      object
RIAA Certification     object
Count Songs             int64
Year                    int64
dtype: object

In [4]:

df = pd.read_csv('beatles.csv', converters={'Released': parse})

In [5]:

df.dtypes

Out[5]:

Unnamed: 0                     int64
Title                         object
Released              datetime64[ns]
Label                         object
UK Chart Position            float64
US Chart Position            float64
BPI Certification             object
RIAA Certification            object
Count Songs                    int64
Year                           int64
dtype: object

In [6]:

dates = ['10-10-2015', '9/8/2012', '2/2014', 'Sep 17, 2000', 'foo']
ser = pd.Series(pd.to_datetime(dates, errors='coerce'))

In [7]:

ser

Out[7]:

0   2015-10-10
1   2012-09-08
2   2014-02-01
3   2000-09-17
4          NaT
dtype: datetime64[ns]

In [8]:

ser.dt

Out[8]:

<pandas.tseries.common.DatetimeProperties object at 0x10f3e49e8>

In [9]:

ser.dt.year

Out[9]:

0    2015.0
1    2012.0
2    2014.0
3    2000.0
4       NaN
dtype: float64

In [10]:

ser.dt.month

Out[10]:

0    10.0
1     9.0
2     2.0
3     9.0
4     NaN
dtype: float64

In [11]:

ser.dt.hour

Out[11]:

0    0.0
1    0.0
2    0.0
3    0.0
4    NaN
dtype: float64

Window Functions¶

In [12]:

idx = pd.date_range('1/1/2000', periods=500, freq='d')

In [13]:

idx

Out[13]:

DatetimeIndex(['2000-01-01', '2000-01-02', '2000-01-03', '2000-01-04',
               '2000-01-05', '2000-01-06', '2000-01-07', '2000-01-08',
               '2000-01-09', '2000-01-10',
               ...
               '2001-05-05', '2001-05-06', '2001-05-07', '2001-05-08',
               '2001-05-09', '2001-05-10', '2001-05-11', '2001-05-12',
               '2001-05-13', '2001-05-14'],
              dtype='datetime64[ns]', length=500, freq='D')

In [14]:

ser = pd.Series(np.random.randn(len(idx)), index=idx)

In [15]:

ser

Out[15]:

2000-01-01   -0.306307
2000-01-02   -0.435708
2000-01-03   -1.535846
2000-01-04    0.409924
2000-01-05   -0.666694
2000-01-06   -0.480828
2000-01-07    0.379973
2000-01-08   -0.773511
2000-01-09    0.396353
2000-01-10   -0.294093
2000-01-11    0.442196
2000-01-12   -1.519349
2000-01-13   -0.649513
2000-01-14   -2.263706
2000-01-15   -0.340966
2000-01-16    0.435463
2000-01-17   -1.666545
2000-01-18    1.452651
2000-01-19   -0.037077
2000-01-20    0.538926
2000-01-21   -0.042326
2000-01-22   -0.227057
2000-01-23    1.024661
2000-01-24   -0.687116
2000-01-25    1.405593
2000-01-26   -0.144300
2000-01-27    0.630102
2000-01-28    0.064640
2000-01-29   -0.376007
2000-01-30    0.017697
                ...   
2001-04-15   -1.672253
2001-04-16   -1.820955
2001-04-17    0.589024
2001-04-18   -0.212713
2001-04-19    2.062938
2001-04-20   -0.318086
2001-04-21   -0.284368
2001-04-22    0.135038
2001-04-23   -0.435896
2001-04-24    1.432767
2001-04-25   -0.775905
2001-04-26    0.086192
2001-04-27    1.469889
2001-04-28   -0.358339
2001-04-29    1.509992
2001-04-30   -0.467628
2001-05-01   -0.810806
2001-05-02    0.352891
2001-05-03   -0.994101
2001-05-04    0.586770
2001-05-05   -0.028301
2001-05-06   -0.308731
2001-05-07   -2.092958
2001-05-08    1.126250
2001-05-09   -0.172683
2001-05-10   -0.826971
2001-05-11    0.324843
2001-05-12    1.402411
2001-05-13    0.371532
2001-05-14   -1.213970
Freq: D, dtype: float64

In [16]:

ser2 = ser + ser.shift(2)

In [17]:

ser2

Out[17]:

2000-01-01         NaN
2000-01-02         NaN
2000-01-03   -1.842153
2000-01-04   -0.025784
2000-01-05   -2.202540
2000-01-06   -0.070904
2000-01-07   -0.286721
2000-01-08   -1.254338
2000-01-09    0.776326
2000-01-10   -1.067604
2000-01-11    0.838549
2000-01-12   -1.813442
2000-01-13   -0.207317
2000-01-14   -3.783055
2000-01-15   -0.990478
2000-01-16   -1.828243
2000-01-17   -2.007510
2000-01-18    1.888114
2000-01-19   -1.703621
2000-01-20    1.991577
2000-01-21   -0.079402
2000-01-22    0.311869
2000-01-23    0.982335
2000-01-24   -0.914173
2000-01-25    2.430253
2000-01-26   -0.831416
2000-01-27    2.035695
2000-01-28   -0.079660
2000-01-29    0.254096
2000-01-30    0.082337
                ...   
2001-04-15   -2.869890
2001-04-16   -2.238086
2001-04-17   -1.083230
2001-04-18   -2.033668
2001-04-19    2.651962
2001-04-20   -0.530799
2001-04-21    1.778571
2001-04-22   -0.183047
2001-04-23   -0.720264
2001-04-24    1.567806
2001-04-25   -1.211801
2001-04-26    1.518959
2001-04-27    0.693983
2001-04-28   -0.272147
2001-04-29    2.979880
2001-04-30   -0.825967
2001-05-01    0.699186
2001-05-02   -0.114737
2001-05-03   -1.804907
2001-05-04    0.939660
2001-05-05   -1.022402
2001-05-06    0.278039
2001-05-07   -2.121259
2001-05-08    0.817519
2001-05-09   -2.265641
2001-05-10    0.299279
2001-05-11    0.152159
2001-05-12    0.575440
2001-05-13    0.696375
2001-05-14    0.188441
Freq: D, dtype: float64

In [18]:

ser.resample('Q').sum()

Out[18]:

2000-03-31     3.521232
2000-06-30     0.306693
2000-09-30    -6.283600
2000-12-31    -0.992922
2001-03-31    16.540240
2001-06-30    -3.598359
Freq: Q-DEC, dtype: float64

In [19]:

ser.resample('Q', label='left').sum()

Out[19]:

1999-12-31     3.521232
2000-03-31     0.306693
2000-06-30    -6.283600
2000-09-30    -0.992922
2000-12-31    16.540240
2001-03-31    -3.598359
Freq: Q-DEC, dtype: float64

In [20]:

ser.resample('m').mean()

Out[20]:

2000-01-31   -0.189758
2000-02-29    0.078981
2000-03-31    0.229460
2000-04-30    0.042248
2000-05-31    0.103755
2000-06-30   -0.139238
2000-07-31   -0.280710
2000-08-31    0.077720
2000-09-30    0.000302
2000-10-31    0.169006
2000-11-30   -0.021901
2000-12-31   -0.179841
2001-01-31   -0.010485
2001-02-28    0.281741
2001-03-31    0.289565
2001-04-30   -0.043818
2001-05-31   -0.163130
Freq: M, dtype: float64

In [21]:

start = datetime.datetime(2010, 1, 1)
end = datetime.datetime(2016, 12, 28)
yahoo_df = data.DataReader('F', 'yahoo', start, end)

In [22]:

yahoo_df.tail().T

Out[22]:

Date	2016-12-21 00:00:00	2016-12-22 00:00:00	2016-12-23 00:00:00	2016-12-27 00:00:00	2016-12-28 00:00:00
Open	12.73	12.63	12.43	12.43	12.37
High	12.77	12.64	12.46	12.51	12.45
Low	12.64	12.40	12.36	12.36	12.22
Close	12.64	12.40	12.46	12.39	12.25
Volume	18056600.00	27556900.00	15578200.00	19403000.00	26678100.00
Adj Close	12.64	12.40	12.46	12.39	12.25

In [23]:

yahoo_df['mean50'] = yahoo_df['Close'].rolling(window=50, center=False).mean()
yahoo_df

Out[23]:

	Open	High	Low	Close	Volume	Adj Close	mean50
Date
2010-01-04	10.17	10.28	10.05	10.28	60855800	8.554412	NaN
2010-01-05	10.45	11.24	10.40	10.96	215620200	9.120268	NaN
2010-01-06	11.21	11.46	11.13	11.37	200070600	9.461446	NaN
2010-01-07	11.46	11.69	11.32	11.66	130201700	9.702767	NaN
2010-01-08	11.67	11.74	11.46	11.69	130463000	9.727731	NaN
2010-01-11	11.90	12.14	11.78	12.11	170626200	10.077230	NaN
2010-01-12	11.98	12.03	11.72	11.87	162995900	9.877516	NaN
2010-01-13	11.91	11.93	11.47	11.68	154527100	9.719410	NaN
2010-01-14	11.65	11.86	11.51	11.76	116531200	9.785981	NaN
2010-01-15	11.74	11.76	11.55	11.60	96149800	9.652838	NaN
2010-01-19	11.51	11.83	11.46	11.75	65934000	9.777659	NaN
2010-01-20	11.68	11.69	11.50	11.51	71649500	9.577946	NaN
2010-01-21	11.53	11.62	11.01	11.18	121451400	9.303339	NaN
2010-01-22	11.01	11.12	10.41	10.52	161530100	8.754126	NaN
2010-01-25	10.73	11.10	10.61	11.03	121621500	9.178517	NaN
2010-01-26	11.17	11.46	11.07	11.19	108250500	9.311660	NaN
2010-01-27	11.57	11.62	11.22	11.55	105091600	9.611231	NaN
2010-01-28	11.90	11.95	11.27	11.41	203320000	9.494731	NaN
2010-01-29	11.60	11.61	10.70	10.84	159741200	9.020411	NaN
2010-02-01	11.14	11.18	10.93	11.12	82748200	9.253410	NaN
2010-02-02	11.26	11.52	11.19	11.39	119714900	9.478089	NaN
2010-02-03	11.49	11.66	11.42	11.64	90125500	9.686124	NaN
2010-02-04	11.49	11.53	11.00	11.06	129792200	9.203482	NaN
2010-02-05	10.97	11.11	10.49	10.91	181535200	9.078661	NaN
2010-02-08	11.09	11.32	10.88	10.97	92031400	9.128589	NaN
2010-02-09	11.18	11.22	11.02	11.15	83207100	9.278374	NaN
2010-02-10	11.12	11.14	10.90	10.94	73395600	9.103625	NaN
2010-02-11	11.00	11.19	10.88	11.18	65116200	9.303339	NaN
2010-02-12	10.92	11.18	10.85	11.12	69465400	9.253410	NaN
2010-02-16	11.21	11.38	11.11	11.32	62537500	9.419838	NaN
...	...	...	...	...	...	...	...
2016-11-15	12.06	12.14	11.95	12.04	31498500	12.040000	12.0164
2016-11-16	12.00	12.05	11.92	12.00	25195500	12.000000	12.0024
2016-11-17	11.90	11.96	11.78	11.87	38502500	11.870000	11.9852
2016-11-18	11.87	11.90	11.73	11.76	30662000	11.760000	11.9728
2016-11-21	11.77	11.90	11.77	11.79	19508600	11.790000	11.9546
2016-11-22	11.76	11.91	11.73	11.89	24274700	11.890000	11.9448
2016-11-23	11.90	12.00	11.85	11.90	22414000	11.900000	11.9400
2016-11-25	11.95	12.06	11.93	12.04	11439600	12.040000	11.9386
2016-11-28	12.02	12.05	11.89	11.92	24407200	11.920000	11.9348
2016-11-29	11.93	12.00	11.87	11.92	23690800	11.920000	11.9310
2016-11-30	11.97	12.14	11.92	11.96	47646200	11.960000	11.9302
2016-12-01	12.23	12.80	12.18	12.43	95143600	12.430000	11.9370
2016-12-02	12.58	12.58	12.15	12.24	38601600	12.240000	11.9382
2016-12-05	12.31	12.50	12.29	12.44	36118300	12.440000	11.9436
2016-12-06	12.40	12.59	12.37	12.56	23668100	12.560000	11.9546
2016-12-07	12.59	13.15	12.54	13.06	58440500	13.060000	11.9762
2016-12-08	13.07	13.15	12.97	13.03	41451800	13.030000	11.9950
2016-12-09	13.00	13.20	12.91	13.17	45233200	13.170000	12.0190
2016-12-12	13.08	13.10	12.81	12.82	39544700	12.820000	12.0340
2016-12-13	12.84	12.88	12.75	12.77	38816300	12.770000	12.0474
2016-12-14	12.65	12.79	12.50	12.53	42243500	12.530000	12.0540
2016-12-15	12.54	12.72	12.52	12.58	23672100	12.580000	12.0560
2016-12-16	12.66	12.70	12.56	12.63	28099200	12.630000	12.0608
2016-12-19	12.59	12.80	12.59	12.66	22780900	12.660000	12.0682
2016-12-20	12.70	12.82	12.68	12.78	17752900	12.780000	12.0814
2016-12-21	12.73	12.77	12.64	12.64	18056600	12.640000	12.0944
2016-12-22	12.63	12.64	12.40	12.40	27556900	12.400000	12.1032
2016-12-23	12.43	12.46	12.36	12.46	15578200	12.460000	12.1142
2016-12-27	12.43	12.51	12.36	12.39	19403000	12.390000	12.1238
2016-12-28	12.37	12.45	12.22	12.25	26678100	12.250000	12.1312

1760 rows × 7 columns

In [24]:

yahoo_df['mean200'] = yahoo_df['Close'].rolling(window=200, center=False).mean()
yahoo_df

Out[24]:

	Open	High	Low	Close	Volume	Adj Close	mean50	mean200
Date
2010-01-04	10.17	10.28	10.05	10.28	60855800	8.554412	NaN	NaN
2010-01-05	10.45	11.24	10.40	10.96	215620200	9.120268	NaN	NaN
2010-01-06	11.21	11.46	11.13	11.37	200070600	9.461446	NaN	NaN
2010-01-07	11.46	11.69	11.32	11.66	130201700	9.702767	NaN	NaN
2010-01-08	11.67	11.74	11.46	11.69	130463000	9.727731	NaN	NaN
2010-01-11	11.90	12.14	11.78	12.11	170626200	10.077230	NaN	NaN
2010-01-12	11.98	12.03	11.72	11.87	162995900	9.877516	NaN	NaN
2010-01-13	11.91	11.93	11.47	11.68	154527100	9.719410	NaN	NaN
2010-01-14	11.65	11.86	11.51	11.76	116531200	9.785981	NaN	NaN
2010-01-15	11.74	11.76	11.55	11.60	96149800	9.652838	NaN	NaN
2010-01-19	11.51	11.83	11.46	11.75	65934000	9.777659	NaN	NaN
2010-01-20	11.68	11.69	11.50	11.51	71649500	9.577946	NaN	NaN
2010-01-21	11.53	11.62	11.01	11.18	121451400	9.303339	NaN	NaN
2010-01-22	11.01	11.12	10.41	10.52	161530100	8.754126	NaN	NaN
2010-01-25	10.73	11.10	10.61	11.03	121621500	9.178517	NaN	NaN
2010-01-26	11.17	11.46	11.07	11.19	108250500	9.311660	NaN	NaN
2010-01-27	11.57	11.62	11.22	11.55	105091600	9.611231	NaN	NaN
2010-01-28	11.90	11.95	11.27	11.41	203320000	9.494731	NaN	NaN
2010-01-29	11.60	11.61	10.70	10.84	159741200	9.020411	NaN	NaN
2010-02-01	11.14	11.18	10.93	11.12	82748200	9.253410	NaN	NaN
2010-02-02	11.26	11.52	11.19	11.39	119714900	9.478089	NaN	NaN
2010-02-03	11.49	11.66	11.42	11.64	90125500	9.686124	NaN	NaN
2010-02-04	11.49	11.53	11.00	11.06	129792200	9.203482	NaN	NaN
2010-02-05	10.97	11.11	10.49	10.91	181535200	9.078661	NaN	NaN
2010-02-08	11.09	11.32	10.88	10.97	92031400	9.128589	NaN	NaN
2010-02-09	11.18	11.22	11.02	11.15	83207100	9.278374	NaN	NaN
2010-02-10	11.12	11.14	10.90	10.94	73395600	9.103625	NaN	NaN
2010-02-11	11.00	11.19	10.88	11.18	65116200	9.303339	NaN	NaN
2010-02-12	10.92	11.18	10.85	11.12	69465400	9.253410	NaN	NaN
2010-02-16	11.21	11.38	11.11	11.32	62537500	9.419838	NaN	NaN
...	...	...	...	...	...	...	...	...
2016-11-15	12.06	12.14	11.95	12.04	31498500	12.040000	12.0164	12.70215
2016-11-16	12.00	12.05	11.92	12.00	25195500	12.000000	12.0024	12.70485
2016-11-17	11.90	11.96	11.78	11.87	38502500	11.870000	11.9852	12.70655
2016-11-18	11.87	11.90	11.73	11.76	30662000	11.760000	11.9728	12.70810
2016-11-21	11.77	11.90	11.77	11.79	19508600	11.790000	11.9546	12.70910
2016-11-22	11.76	11.91	11.73	11.89	24274700	11.890000	11.9448	12.71180
2016-11-23	11.90	12.00	11.85	11.90	22414000	11.900000	11.9400	12.71460
2016-11-25	11.95	12.06	11.93	12.04	11439600	12.040000	11.9386	12.71895
2016-11-28	12.02	12.05	11.89	11.92	24407200	11.920000	11.9348	12.72080
2016-11-29	11.93	12.00	11.87	11.92	23690800	11.920000	11.9310	12.72105
2016-11-30	11.97	12.14	11.92	11.96	47646200	11.960000	11.9302	12.71950
2016-12-01	12.23	12.80	12.18	12.43	95143600	12.430000	11.9370	12.72035
2016-12-02	12.58	12.58	12.15	12.24	38601600	12.240000	11.9382	12.72105
2016-12-05	12.31	12.50	12.29	12.44	36118300	12.440000	11.9436	12.72045
2016-12-06	12.40	12.59	12.37	12.56	23668100	12.560000	11.9546	12.72115
2016-12-07	12.59	13.15	12.54	13.06	58440500	13.060000	11.9762	12.72605
2016-12-08	13.07	13.15	12.97	13.03	41451800	13.030000	11.9950	12.72925
2016-12-09	13.00	13.20	12.91	13.17	45233200	13.170000	12.0190	12.73275
2016-12-12	13.08	13.10	12.81	12.82	39544700	12.820000	12.0340	12.73430
2016-12-13	12.84	12.88	12.75	12.77	38816300	12.770000	12.0474	12.73270
2016-12-14	12.65	12.79	12.50	12.53	42243500	12.530000	12.0540	12.72925
2016-12-15	12.54	12.72	12.52	12.58	23672100	12.580000	12.0560	12.72445
2016-12-16	12.66	12.70	12.56	12.63	28099200	12.630000	12.0608	12.71965
2016-12-19	12.59	12.80	12.59	12.66	22780900	12.660000	12.0682	12.71485
2016-12-20	12.70	12.82	12.68	12.78	17752900	12.780000	12.0814	12.71260
2016-12-21	12.73	12.77	12.64	12.64	18056600	12.640000	12.0944	12.70985
2016-12-22	12.63	12.64	12.40	12.40	27556900	12.400000	12.1032	12.70610
2016-12-23	12.43	12.46	12.36	12.46	15578200	12.460000	12.1142	12.70195
2016-12-27	12.43	12.51	12.36	12.39	19403000	12.390000	12.1238	12.69755
2016-12-28	12.37	12.45	12.22	12.25	26678100	12.250000	12.1312	12.69265

1760 rows × 8 columns

In [25]:

%matplotlib inline
yahoo_df.plot()

Out[25]:

<matplotlib.axes._subplots.AxesSubplot at 0x10ffab780>

In [26]:

yahoo_df[['Close', 'mean50', 'mean200']].plot()

Out[26]:

<matplotlib.axes._subplots.AxesSubplot at 0x11013e9e8>

In [ ]: