ContextQMD
Back to Influxdb

statsmodels.linearRegression() function

content/flux/v0/stdlib/contrib/anaisdg/statsmodels/linearregression.md

latest2.1 KB
Original Source
<!------------------------------------------------------------------------------ IMPORTANT: This page was generated from comments in the Flux source code. Any edits made directly to this page will be overwritten the next time the documentation is generated. To make updates to this documentation, update the function comments above the function definition in the Flux source code: https://github.com/influxdata/flux/blob/master/stdlib/contrib/anaisdg/statsmodels/linearreg.flux#L45-L93 Contributing to Flux: https://github.com/influxdata/flux#contributing Fluxdoc syntax: https://github.com/influxdata/flux/blob/master/docs/fluxdoc.md ------------------------------------------------------------------------------->

statsmodels.linearRegression() performs a linear regression.

It calculates and returns ŷ (y_hat), and residual sum of errors (rse). Output data includes the following columns:

  • N: Number of points in the calculation.
  • slope: Slope of the calculated regression.
  • sx: Sum of x.
  • sxx: Sum of x squared.
  • sxy: Sum of x*y.
  • sy: Sum of y.
  • errors: Residual sum of squares. Defined by (r.y - r.y_hat) ^ 2 in this context
  • x: An index [1,2,3,4...n], with the assumption that the timestamps are regularly spaced.
  • y: Field value
  • y_hat: Linear regression values
Function type signature
js
(
    <-tables: stream[A],
) => stream[{
    B with
    y_hat: float,
    y: float,
    x: float,
    sy: H,
    sxy: G,
    sxx: F,
    sx: E,
    slope: D,
    errors: float,
    N: C,
}] where A: Record, D: Divisible + Subtractable

{{% caption %}} For more information, see Function type signatures. {{% /caption %}}

Parameters

tables

Input data. Default is piped-forward data (<-).

Examples

Perform a linear regression on a dataset

js
import "contrib/anaisdg/statsmodels"
import "sampledata"

sampledata.float()
    |> statsmodels.linearRegression()