Robust Locally Weighted Regression and Smoothing Scatterplots

By William S. Cleveland. 1979

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Local Linear Forests (2018, methodology): “Cleveland (1979) expanded on this by introducing robust locally weighted regression, and Fan and Gijbels (1992) gave a variable bandwidth version.”
Spatial Dependence in Local Unemployment Rates (2007, methodology): “ui,t = αui,t−1 + βui,t−1 + γθi,t−1 + δθi,t−1 + ηi + εi,t, (9)”
The utility of L-moment ratio diagrams for selecting a regional probability distribution (2001, methodology): “The modification to the LOWESS of Cleveland (1979) takes into account the variance in both L-skewness and L-kurtosis, not just L-kurtosis.”
The ATLAS3D project – XX. Mass–size and mass–σ distributions of early-type galaxies: bulge fraction drives kinematics, mass-to-light ratio, molecular gas fraction and stellar initial mass function (2012, methodology): “1, we overplot a version of the MJAM–σe relation which was adaptively smoothed using the onedimensional locally weighted regression robust technique (dubbed LOESS3) of Cleveland (1979).”
Modelling Palaeoecological Time Series Using Generalised Additive Models (2018, methodology): “LOESS, locally weighted scatterplot smoother, as it’s name suggests, was developed to smooth x-y scatterplot data (Cleveland, 1979).”
Nonparametric Identification in Nonseparable Panel Data Models with Generalized Fixed Effects (2009, methodology): “21”
The morphological family size effect and morphology (2000, background): “The scatterplots shown in Figure 2 do not suggest severe violations of homoscedasticity and the non-parametric regression smoothers (Cleveland, 1979) likewise suggest roughly linear trends.”
Analysis of air quality time series of Hong Kong with graphical modeling (2016, methodology): “In each figure, lines constructed from the LOESS smoother (Cleveland, 1979) are also shown.”
Using mixed models to combine smooth size-selection and catch-comparison curves over hauls (2003, methodology): “J:\cjfas\cjfas60\cjfas6004\F03-029.vp Wednesday, May 28, 2003 6:57:22 AM kernel smoothers (Hastie and Tibshirani 1990) and the loess smoother of Cleveland (1979).”
Local linear spatial regression (2004, background): “…documented [see, e.g., Fan and Gijbels (1996)] that this approach suffers from several severe drawbacks, such as poor boundary performance, excessive bias and low efficiency, and that the local polynomial fitting methods developed by Stone (1977) and Cleveland (1979) are generally preferable.”
Generalized Additive Models (1986, unknown): “called scatterplot smoother, for example a running mean, running median, running least squares line, kernel estimate, or a spline (see Reinsch (1967), Wahba and Wold (1975), Cleveland (1979), and Silverman (1985) for discussions of smoothing techniques).”
Rethinking Parameter Counting in Deep Models: Effective Dimensionality Revisited (2020, methodology): “Cleveland (1979) introduced effective dimensionality into the splines literature as a measure of goodness of ﬁt, while Hastie & Tibshirani (1990, Chapter 3) used it to assess generalized additive models.”
Smallholder Income and Land Distribution in Africa: Implications for Poverty Reduction Strategies (2003, unknown): “…lines show the 25 th , 50 th , and 75 th percentiles The locally-weighted regression lines in Figure 1 are created using locally weighted smoothed scatter plots (LOWESS) with window length set at .6 or .7 of the neighboring observations (Cleveland 1979). of sampled households along the x-axis.”
Referential kNN Regression for Financial Time Series Forecasting (2013, methodology): “The application of kNN to time series forecasting under nonparametric locally weighted regression condition was presented independently by Yakowitz [14] and Cleveland [15] within the community of statistics.”