# Cross validation, n-fold and leave-one-out for generalized least squares ('gls') This function is a cross validation function for 'gls' method in 'nlme' package. ```r glscv( model = var1 ~ 1, trainxy, y, corr.args = NULL, weights = NULL, validation = "CV", cv.fold = 10, predacc = "VEcv", ... ) ``` ## Arguments - `model`: a formula defining the response variable and predictive variables. - `trainxy`: a dataframe contains longitude (long), latitude (lat), predictive variables and the response variable of point samples. That is, the location information must be names as 'long' and 'lat'. - `y`: a vector of the response variable in the formula, that is, the left part of the formula. - `corr.args`: arguments for 'correlation' in 'gls'. See '?corClasses' in 'nlme' for details. By default, "NULL" is used. When "NULL" is used, then 'gls' is actually performing 'lm'. - `weights`: describing the within-group heteroscedasticity structure. Defaults to "NULL", corresponding to homoscedastic errors. See '?gls' in 'nlme' for details. - `validation`: validation methods, include 'LOO': leave-one-out, and 'CV': cross-validation. - `cv.fold`: integer; number of folds in the cross-validation. if > 1, then apply n-fold cross validation; the default is 10, i.e., 10-fold cross validation that is recommended. - `predacc`: can be either "VEcv" for vecv or "ALL" for all measures in function pred.acc. - `...`: other arguments passed on to 'gls'. ## Returns A list with the following components: me, rme, mae, rmae, mse, rmse, rrmse, vecv and e1; or vecv only ## Note This function is largely based on rfcv in 'randomForest' and 'gls' in 'library(nlme)'. ## Examples ```r library(spm) library(nlme) data(petrel) gravel <- petrel[, c(1, 2, 6:9, 5)] range1 <- 0.8 nugget1 <- 0.5 model <- log(gravel + 1) ~ long + lat + bathy + dist + I(long^2) + I(lat^2) + I(lat^3) + I(bathy^2) + I(bathy^3) + I(dist^2) + I(dist^3) + I(relief^2) + I(relief^3) glscv1 <- glscv(model = model, gravel, log(gravel[, 7] +1), validation = "CV", corr.args = corSpher(c(range1, nugget1), form = ~ long + lat, nugget = TRUE), predacc = "ALL") glscv1 #For gls set.seed(1234) n <- 20 # number of iterations,60 to 100 is recommended. VEcv <- NULL for (i in 1:n) { glscv1 <- glscv(model = model, gravel, log(gravel[, 7] +1), validation = "CV", corr.args = corSpher(c(range1, nugget1), form = ~ long + lat, nugget = TRUE), predacc = "VEcv") VEcv [i] <- glscv1 } plot(VEcv ~ c(1:n), xlab = "Iteration for GLS", ylab = "VEcv (%)") points(cumsum(VEcv) / c(1:n) ~ c(1:n), col = 2) abline(h = mean(VEcv), col = 'blue', lwd = 2) # For lm, that is, gls with 'correlation = NULL' n <- 20 # number of iterations,60 to 100 is recommended. VEcv <- NULL set.seed(1234) for (i in 1:n) { glscv1 <- glscv(model = model, gravel, log(gravel[, 7] +1), validation = "CV", predacc = "VEcv") VEcv [i] <- glscv1 } plot(VEcv ~ c(1:n), xlab = "Iteration for GLS", ylab = "VEcv (%)") points(cumsum(VEcv) / c(1:n) ~ c(1:n), col = 2) abline(h = mean(VEcv), col = 'blue', lwd = 2) ``` ## References Pinheiro, J. C. and D. M. Bates (2000). Mixed-Effects Models in S and S-PLUS. New York, Springer. ## Author(s) Jin Li

glscv function