naginterfaces.library.smooth.data_order(x, y, wt=None)[source]

data_order orders and weights data which is entered unsequentially, weighted or unweighted.

For full information please refer to the NAG Library document for g10za

xfloat, array-like, shape

The values, , for .

yfloat, array-like, shape

The values , for .

wtNone or float, array-like, shape , optional

If , must contain the weights. Otherwise is not referenced and unit weights are assumed.


The number of distinct observations.

xordfloat, ndarray, shape

The first elements contain the ordered and distinct .

yordfloat, ndarray, shape

The first elements contain the values corresponding to the values in .

wtordfloat, ndarray, shape

The first elements contain the values corresponding to the values of and .


The within group sum of squares for tied observations.

(errno )

On entry, .

Constraint: or .

(errno )

On entry, .

Constraint: .

(errno )

On entry, all weights are zero.

(errno )

On entry, and .

Constraint: .


In the NAG Library the traditional C interface for this routine uses a different algorithmic base. Please contact NAG if you have any questions about compatibility.

Given a set of observations , for , with corresponding weights , data_order rearranges the observations so that the are in ascending order.

For any equal in the ordered set, say , a single observation is returned with a corresponding and , calculated as


Observations with zero weight are ignored. If no weights are supplied by you, then unit weights are assumed; that is , for .

In addition, the within group sum of squares is computed for the tied observations using West’s algorithm (see West (1979)).


Draper, N R and Smith, H, 1985, Applied Regression Analysis, (2nd Edition), Wiley

West, D H D, 1979, Updating mean and variance estimates: An improved method, Comm. ACM (22), 532–555