@@ -2897,18 +2897,21 @@ def interpolate(self, method='linear', axis=0, limit=None, inplace=False,
2897
2897
'polynomial', 'spline' 'piecewise_polynomial', 'pchip'}
2898
2898
2899
2899
* 'linear': ignore the index and treat the values as equally
2900
- spaced. default
2900
+ spaced. This is the only method supported on MultiIndexes.
2901
+ default
2901
2902
* 'time': interpolation works on daily and higher resolution
2902
2903
data to interpolate given length of interval
2903
2904
* 'index', 'values': use the actual numerical values of the index
2904
2905
* 'nearest', 'zero', 'slinear', 'quadratic', 'cubic',
2905
2906
'barycentric', 'polynomial' is passed to
2906
- `scipy.interpolate.interp1d` with the order given both
2907
+ `scipy.interpolate.interp1d` with the order given. Both
2907
2908
'polynomial' and 'spline' requre that you also specify and order
2908
- (int) e.g. df.interpolate(method='polynomial', order=4)
2909
+ (int) e.g. df.interpolate(method='polynomial', order=4). These
2910
+ use the actual numerical values of the index
2909
2911
* 'krogh', 'piecewise_polynomial', 'spline', and 'pchip' are all
2910
2912
wrappers around the scipy interpolation methods of similar
2911
- names. See the scipy documentation for more on their behavior:
2913
+ names. These use the actual numerical values of the index. See
2914
+ the scipy documentation for more on their behavior:
2912
2915
http://docs.scipy.org/doc/scipy/reference/interpolate.html#univariate-interpolation
2913
2916
http://docs.scipy.org/doc/scipy/reference/tutorial/interpolate.html
2914
2917
0 commit comments