detrend

Subtract mean or best-fit line fromtimeseriesobject

Syntax

tsout = detrend(tsin,method)

tsout = detrend(tsin,method,ind)

Description

tsout = detrend(tsin,method)subtracts either a mean or a best-fit line from 2-D data in atimeseriesusing the specifiedmethod, and also removes allNaNvalues.

example

tsout = detrend(tsin,method,ind)specifies the indices of the columns or rows to detrend.indis a vector of integers representing column indices for column-oriented data (tsin.IsTimeFirstistrue) and representing row indices for row-oriented data (tsin.IsTimeFirstisfalse).

Examples

collapse all

Subtract Mean from Data

Open Live Script

创建一个timeseriesobject with 2-D data, and subtract the mean of each column from the data.

A = magic(3)

A =3×38 1 6 3 5 7 4 9 2

tsin = timeseries(A,[1 2 3]); tsout = detrend(tsin,'constant'); tsout.Data

ans =3×33 -4 1 -2 0 2 -1 4 -3

Subtract the mean of each column for only the second and third columns of data.

tscol = detrend(tsin,'constant',[2 3]); tscol.Data

ans =3×38 -4 1 3 0 2 4 4 -3

Input Arguments

collapse all

`tsin`—Input`timeseries`
scalar

Inputtimeseriescontaining 2-D data, specified as a scalar.

Data Types:timeseries

`method`—Detrend method
`'constant'`|`'linear'`

Detrend method, specified as one of the following options:

'constant'— Subtract the mean from the data.
'linear'— Subtract the best-fit line from the data.

`ind`—Row or column indices
scalar|vector

Row or column indices, specified as a positive integer numeric scalar or vector.indrepresents column indices for column-oriented data (tsin.IsTimeFirstistrue) and represents row indices for row-oriented data (tsin.IsTimeFirstisfalse).

版本History

Introduced before R2006a