我的书签
添加书签
移除书签聚合
Once the Rolling, Expanding or EWM objects have been created, several methods are available to perform multiple computations on the data. These operations are similar to the aggregating API, groupby API, and resample API.
In [85]: dfa = pd.DataFrame(np.random.randn(1000, 3),....: index=pd.date_range('1/1/2000', periods=1000),....: columns=['A', 'B', 'C'])....:In [86]: r = dfa.rolling(window=60,min_periods=1)In [87]: rOut[87]: Rolling [window=60,min_periods=1,center=False,axis=0]
We can aggregate by passing a function to the entire DataFrame, or select a Series (or multiple Series) via standard __getitem__.
In [88]: r.aggregate(np.sum)Out[88]:A B C2000-01-01 -0.289838 -0.370545 -1.2842062000-01-02 -0.216612 -1.675528 -1.1694152000-01-03 1.154661 -1.634017 -1.5666202000-01-04 2.969393 -4.003274 -1.8161792000-01-05 4.690630 -4.682017 -2.7172092000-01-06 3.880630 -4.447700 -1.0789472000-01-07 4.001957 -2.884072 -3.116903... ... ... ...2002-09-20 2.652493 -10.528875 9.8678052002-09-21 0.844497 -9.280944 9.5226492002-09-22 2.860036 -9.270337 6.4152452002-09-23 3.510163 -8.151439 5.1772192002-09-24 6.524983 -10.168078 5.7926392002-09-25 6.409626 -9.956226 5.7040502002-09-26 5.093787 -7.074515 6.905823[1000 rows x 3 columns]In [89]: r['A'].aggregate(np.sum)Out[89]:2000-01-01 -0.2898382000-01-02 -0.2166122000-01-03 1.1546612000-01-04 2.9693932000-01-05 4.6906302000-01-06 3.8806302000-01-07 4.001957...2002-09-20 2.6524932002-09-21 0.8444972002-09-22 2.8600362002-09-23 3.5101632002-09-24 6.5249832002-09-25 6.4096262002-09-26 5.093787Freq: D, Name: A, Length: 1000, dtype: float64In [90]: r[['A','B']].aggregate(np.sum)Out[90]:A B2000-01-01 -0.289838 -0.3705452000-01-02 -0.216612 -1.6755282000-01-03 1.154661 -1.6340172000-01-04 2.969393 -4.0032742000-01-05 4.690630 -4.6820172000-01-06 3.880630 -4.4477002000-01-07 4.001957 -2.884072... ... ...2002-09-20 2.652493 -10.5288752002-09-21 0.844497 -9.2809442002-09-22 2.860036 -9.2703372002-09-23 3.510163 -8.1514392002-09-24 6.524983 -10.1680782002-09-25 6.409626 -9.9562262002-09-26 5.093787 -7.074515[1000 rows x 2 columns]
As you can see, the result of the aggregation will have the selected columns, or all columns if none are selected.
Applying multiple functions
With windowed Series you can also pass a list of functions to do aggregation with, outputting a DataFrame:
In [91]: r['A'].agg([np.sum, np.mean, np.std])Out[91]:sum mean std2000-01-01 -0.289838 -0.289838 NaN2000-01-02 -0.216612 -0.108306 0.2567252000-01-03 1.154661 0.384887 0.8733112000-01-04 2.969393 0.742348 1.0097342000-01-05 4.690630 0.938126 0.9779142000-01-06 3.880630 0.646772 1.1288832000-01-07 4.001957 0.571708 1.049487... ... ... ...2002-09-20 2.652493 0.044208 1.1649192002-09-21 0.844497 0.014075 1.1482312002-09-22 2.860036 0.047667 1.1320512002-09-23 3.510163 0.058503 1.1342962002-09-24 6.524983 0.108750 1.1442042002-09-25 6.409626 0.106827 1.1429132002-09-26 5.093787 0.084896 1.151416[1000 rows x 3 columns]
On a windowed DataFrame, you can pass a list of functions to apply to each column, which produces an aggregated result with a hierarchical index:
In [92]: r.agg([np.sum, np.mean])Out[92]:A B Csum mean sum mean sum mean2000-01-01 -0.289838 -0.289838 -0.370545 -0.370545 -1.284206 -1.2842062000-01-02 -0.216612 -0.108306 -1.675528 -0.837764 -1.169415 -0.5847082000-01-03 1.154661 0.384887 -1.634017 -0.544672 -1.566620 -0.5222072000-01-04 2.969393 0.742348 -4.003274 -1.000819 -1.816179 -0.4540452000-01-05 4.690630 0.938126 -4.682017 -0.936403 -2.717209 -0.5434422000-01-06 3.880630 0.646772 -4.447700 -0.741283 -1.078947 -0.1798252000-01-07 4.001957 0.571708 -2.884072 -0.412010 -3.116903 -0.445272... ... ... ... ... ... ...2002-09-20 2.652493 0.044208 -10.528875 -0.175481 9.867805 0.1644632002-09-21 0.844497 0.014075 -9.280944 -0.154682 9.522649 0.1587112002-09-22 2.860036 0.047667 -9.270337 -0.154506 6.415245 0.1069212002-09-23 3.510163 0.058503 -8.151439 -0.135857 5.177219 0.0862872002-09-24 6.524983 0.108750 -10.168078 -0.169468 5.792639 0.0965442002-09-25 6.409626 0.106827 -9.956226 -0.165937 5.704050 0.0950682002-09-26 5.093787 0.084896 -7.074515 -0.117909 6.905823 0.115097[1000 rows x 6 columns]
Passing a dict of functions has different behavior by default, see the next section.
Applying different functions to DataFrame columns
By passing a dict to aggregate you can apply a different aggregation to the columns of a DataFrame:
In [93]: r.agg({'A' : np.sum,....: 'B' : lambda x: np.std(x, ddof=1)})....:Out[93]:A B2000-01-01 -0.289838 NaN2000-01-02 -0.216612 0.6607472000-01-03 1.154661 0.6899292000-01-04 2.969393 1.0721992000-01-05 4.690630 0.9396572000-01-06 3.880630 0.9668482000-01-07 4.001957 1.240137... ... ...2002-09-20 2.652493 1.1148142002-09-21 0.844497 1.1132202002-09-22 2.860036 1.1132082002-09-23 3.510163 1.1323812002-09-24 6.524983 1.0809632002-09-25 6.409626 1.0829112002-09-26 5.093787 1.136199[1000 rows x 2 columns]
The function names can also be strings. In order for a string to be valid it must be implemented on the windowed object
In [94]: r.agg({'A' : 'sum', 'B' : 'std'})Out[94]:A B2000-01-01 -0.289838 NaN2000-01-02 -0.216612 0.6607472000-01-03 1.154661 0.6899292000-01-04 2.969393 1.0721992000-01-05 4.690630 0.9396572000-01-06 3.880630 0.9668482000-01-07 4.001957 1.240137... ... ...2002-09-20 2.652493 1.1148142002-09-21 0.844497 1.1132202002-09-22 2.860036 1.1132082002-09-23 3.510163 1.1323812002-09-24 6.524983 1.0809632002-09-25 6.409626 1.0829112002-09-26 5.093787 1.136199[1000 rows x 2 columns]
Furthermore you can pass a nested dict to indicate different aggregations on different columns.
In [95]: r.agg({'A' : ['sum','std'], 'B' : ['mean','std'] })Out[95]:A Bsum std mean std2000-01-01 -0.289838 NaN -0.370545 NaN2000-01-02 -0.216612 0.256725 -0.837764 0.6607472000-01-03 1.154661 0.873311 -0.544672 0.6899292000-01-04 2.969393 1.009734 -1.000819 1.0721992000-01-05 4.690630 0.977914 -0.936403 0.9396572000-01-06 3.880630 1.128883 -0.741283 0.9668482000-01-07 4.001957 1.049487 -0.412010 1.240137... ... ... ... ...2002-09-20 2.652493 1.164919 -0.175481 1.1148142002-09-21 0.844497 1.148231 -0.154682 1.1132202002-09-22 2.860036 1.132051 -0.154506 1.1132082002-09-23 3.510163 1.134296 -0.135857 1.1323812002-09-24 6.524983 1.144204 -0.169468 1.0809632002-09-25 6.409626 1.142913 -0.165937 1.0829112002-09-26 5.093787 1.151416 -0.117909 1.136199[1000 rows x 4 columns]
