diagnostic_plots: diagnostic_plots

In ericward-noaa/vista: Visual Inspection of Spatiotemporal Analyses

diagnostic_plots

Description

diagnostic_plots

Usage

diagnostic_plots(
  df,
  X = "X",
  Y = "Y",
  time = "time",
  pred = "pred",
  resid = "resid",
  demean_time = TRUE
)

Arguments

df	Dataframe, containing locations, time, and predictions
X	Character string containing the name of X variable
Y	Character string containing the name of Y variable
time	Character string containing the name of time variable
pred	Character string containing the name of prediction variable
resid	Character string containing the name of residual
demean_time	Boolean, whether or not to remove temporal means (similar to fixed time effects). Defaults to TRUE

Value

A list of ggplot objects that can be manipulated further

Examples

set.seed(2021)
d <- data.frame(
  X = runif(1000), Y = runif(1000),
  year = sample(1:10, size = 1000, replace = TRUE)
)
d$density <- rnorm(0.01 * d$X - 0.001 * d$X * d$X + d$Y * 0.02 - 0.005 * d$Y * d$Y, 0, 0.1)
m <- mgcv::gam(density ~ 0 + as.factor(year) + s(X, Y), data = d)
d$pred <- predict(m)
#  d$resid = residuals(m)
#  # the default names match, with the exception of year -- so change it
#  plots <- diagnostic_plots(d, time="year")

ericward-noaa/vista documentation built on June 24, 2022, 12:19 a.m.