pandas - how to use the 'mean' function
s
the mean of a data frame.
I import stock price data from yahoo for various stock tickers, truncate
the data to be in the desired date range and then I want to compute
various statistics. One of them is the mean.
I have read the documentation on
pandas.Series.mean
pandas.DataFrame.mean
http://pandas.sourceforge.net/generated/pandas.DataFrame.mean.html#pandas.DataFrame.mean
The only problem is I don't know how to call these functions. I can't
seem to import them via
from pandas.series import mean
or
from pandas.dataframe import mean
And when I look in the directory structure there is no such thing as
'series' as a module within pandas.
Apologies if this is a trivial question as I am very new to python and
pandas.
Wes McKinney
Anytime you see something like Series.mean or DataFrame.mean in the
API it means that those are "instance methods", i.e. they are defined
within a class and they have access to the state (data) in the object:
http://docs.python.org/tutorial/classes.html#method-objects
So if you have a DataFrame df, you would do:
df.mean()
to compute the means of the columns. Or
df.mean(1) to compute the means of the rows.
In R when you do:
f(df)
typically what happens under the hood is the R interpreter actually does
f.data.frame(df)
this is similar to instance methods in Python, since
df.mean()
is just syntactic sugar for
DataFrame.mean(df)
but you would never need to know that in 99% of practical use.
I very strongly recommend that you use IPython (ipython.org) as it
makes exploring object methods and there documentation very easy,
e.g.:
In [3]: df.<TAB>
df.A df.iteritems
df.add df.iterkv
df.add_prefix df.ix
df.add_suffix df.join
df.align df.last_valid_index
df.append df.load
df.apply df.mad
df.applymap df.max
df.asfreq df.mean
df.as_matrix df.median
df.astype df.min
df.axes df.mul
df.B df.ndim
df.boxplot df.pivot
df.C df.pivot_table
df.clip df.plot
df.clip_lower df.pop
df.clip_upper df.prod
df.columns df.product
df.combine df.put_value
df.combineAdd df.quantile
df.combine_first df.radd
df.combineMult df.rdiv
df.consolidate df.reindex
df.copy df.reindex_like
df.corr df.rename
df.corrwith df.rename_axis
df.count df.rmul
df.cov df.rsub
df.cumprod df.save
df.cumsum df.select
df.D df.set_index
df.delevel df.shape
df.describe df.shift
df.diff df.skew
df.div df.sort
df.dot df.sort_index
df.drop df.sortlevel
df.drop_duplicates df.stack
df.dropna df.std
df.dtypes df.sub
df.duplicated df.sum
df.fillna df.swaplevel
df.filter df.T
df.first_valid_index df.tail
df.from_csv df.take
df.from_dict df.to_csv
df.from_records df.to_dict
df.get_dtype_counts df.to_html
df.get_value df.to_records
df.groupby df.to_sparse
df.head df.to_string
df.hist df.transpose
df.idxmax df.truncate
df.idxmin df.unstack
df.index df.values
df.info df.var
df.insert df.xs
In [3]: df.mean?
Type: instancemethod
Base Class: <type 'instancemethod'>
String Form:
<bound method DataFrame.mean of A B C D
2000-01-03 -0.770 <...>
2000-02-10 0.9349 -1.102 -0.229 0.824
2000-02-11 0.5129 -0.1323 0.06714 0.231 >
Namespace: Interactive
File: /home/wesm/code/pandas/pandas/core/frame.py
Definition: df.mean(self, axis=0, skipna=True, level=None)
Docstring:
Return mean over requested axis.
NA/null values are excluded
Parameters
----------
axis : {0, 1}
0 for row-wise, 1 for column-wise
skipna : boolean, default True
Exclude NA/null values. If an entire row/column is NA, the result
will be NA
level : int, default None
If the axis is a MultiIndex (hierarchical), count along a
particular level, collapsing into a DataFrame
Returns
-------
mean : Series (or DataFrame if level specified)
- Wes