spaloess: Spatially Local Polynomial Regression Fitting
In XiaosuTong/Spaloess: Spatially Local Polynomial Regression

Description Usage Arguments Details Author(s) Examples

The first layer of the Spatial locally weighted regression, using local fitting with different type of distance calculation.

spaloess(formula, data, weights, subset, na.action, model = FALSE,
  napred = TRUE, span = 0.75, enp.target, degree = 2L,
  parametric = FALSE, distance = "Latlong", alltree = FALSE,
  drop_square = FALSE, normalize = FALSE, family = c("gaussian",
  "symmetric"), method = c("loess", "model.frame"),
  control = loess.control(...), ...)

`formula`	a formula specifying the numeric response and one to four numeric predictors.
`data`	an optional data from, list or environment containing the variables in the model. If not found in 'data', the variables are taken from 'environment', typically the environment from which 'loess' is called.
`weights`	optional weights for each case
`subset`	an optional specification of a subset of the data to be used
`na.action`	the action to be taken with missing values in the response or predictors. The default is given by 'getOption("na.action")'.
`model`	Should the model frame be returned?
`napred`	Should missing observations in the dataset be predicted. Default is TRUE.
`span`	The parameter alpha which controls the portion of data points used in the local fit.
`enp.target`	An alternative way to specify 'span', as the approximate equivalent number of parameters to be used.
`degree`	The degree of the polynomials to be used, normally 1 or 2. (Degree 0 is also allowed, but see the 'Note'.)
`parametric`	should any terms be fitted globally rather than locally? Terms can be specified by name, number or as a logical vector of the same length as the number of predictors.
`distance`	Options: "Euclid", or "Latlong" which is for great circle distance
`alltree`	Should the kd-tree built based on all observations or only non-NA observations.
`drop_square`	For fits with more than one predictor and 'degree = 2', should the quadratic term be dropped for particular predictors? Terms are specified in the same way as for 'parametric'.
`normalize`	Should the predictors be normalized to a common scale if there is more than one? The normalization used is to set the 10 "Latlong" distance.
`family`	If 'gaussian' fitting is by least-squares, and if 'symmetric' a re-descending M estimator is used with Tukey's bi-weight function.
`method`	Fit the model or just extract the model frame.
`control`	control parameters: see 'loess.control'.
`...`	arguments passed to or from other methods.

This spaloess function is the first wrapper of the spatial loess fitting procedure. It checks all the validity of all input arguments, and formats arguments like drop_square, parametric. Also generate other important arguments, like iteration, and pass all arguments into the second wrapper function: newsimpleLoess

Xiaosu Tong, based on 'loess' function of B. D. Ripley, and 'cloess' package of Cleveland, Grosse and Shyu.

    set.seed(66)
    x1 <- rnorm(100, mean=-100, sd=10)
    x2 <- rnorm(100, mean=38, sd=4)
    y <- 0.1*x1 + 1*x2 - 10 + rnorm(100, 0, 1.3)
    testdata <- data.frame(LON = x1, LAT = x2, tmax = y)
    cars.lo <- spaloess(tmax ~ LON + LAT, testdata, distance = "Latlong", napred = FALSE)