Degree of Local Polynomials
The local polynomials fit to each subset of the data are almost always of first or second degree; that is, either locally linear (in the straight line sense) or locally quadratic. Using a zero degree polynomial turns LOESS into a weighted moving average. Such a simple local model might work well for some situations, but may not always approximate the underlying function well enough. Higher-degree polynomials would work in theory, but yield models that are not really in the spirit of LOESS. LOESS is based on the ideas that any function can be well approximated in a small neighborhood by a low-order polynomial and that simple models can be fit to data easily. High-degree polynomials would tend to overfit the data in each subset and are numerically unstable, making accurate computations difficult.
Read more about this topic: Local Regression
Famous quotes containing the words degree of, degree and/or local:
“It has become a people’s war, and peoples of all sorts and races, of every degree of power and variety of fortune, are involved in its sweeping processes of change and settlement.”
—Woodrow Wilson (1856–1924)
“It is a secret from nobody that the famous random event is most likely to arise from those parts of the world where the old adage “There is no alternative to victory” retains a high degree of plausibility.”
—Hannah Arendt (1906–1975)
“While it may not heighten our sympathy, wit widens our horizons by its flashes, revealing remote hidden affiliations and drawing laughter from far afield; humor, in contrast, strikes up fellow feeling, and though it does not leap so much across time and space, enriches our insight into the universal in familiar things, lending it a local habitation and a name.”
—Marie Collins Swabey. Comic Laughter, ch. 5, Yale University Press (1961)