Scatter diagram smoother
As a scatterplot smoothing , scatterplot smoothers or smoothing of scatter plots is called in the statistics , nonparametric approaches that require a flexible modeling of the effect of a continuous covariates on a continuous dependent variable permit.
description
Smoothing tries to separate the random fluctuations in the data from the non- random behavior. The fluctuations can then be removed or reduced in order to make the systematic or non-random relationship between the variables (better) visible.
Two-dimensional scatter plots can be smoothed by fitting a regression curve to the data points on the plot. This curve attempts to indicate the non-random component that the random relationship between the variables is superimposed on.
Usually smoothing is done using one of the following techniques:
- linear single regression (works with regression lines)
- quadratic or polynomial regression (works with polynomial curves)
- local regression
- Spline smoothing.
The smoothing curve is chosen so that, in some way, the best fit is achieved; This is often defined as the adaptation in which the residual sum of squares is minimized ( least squares criterion ).
Individual evidence
- ↑ Ludwig Fahrmeir , Thomas Kneib , Stefan Lang: Regression: Models, Methods and Applications. Springer Verlag, Berlin / Heidelberg / New York 2007, ISBN 978-3-540-33932-8 , p. 292
- ^ Dodge, Y. (2006) The Oxford Dictionary of Statistical Terms , OUP. (Entry for smoothing ″)