Learn R Programming

hsdar (version 0.5.1)

plot.Nri: Plot function for (g)lm.nri and cor.test.nri

Description

Plot values in (generalised) linear modes and correlation tests from narrow band indices

Usage

"plot"(x, coefficient = "p.value", predictor = 2, xlab = "Wavelength band 1 (nm)", ylab = "Wavelength band 2 (nm)", legend = TRUE, colspace = "hcl", col = c(10, 90, 60, 60, 10, 80), digits = 2, range = "auto", constraint  = NULL, uppertriang = FALSE, ...)

Arguments

x
Object to be plotted.
coefficient
Name or index of coefficient to plot.
predictor
Name or index of term to plot.
xlab
Label for x-axis.
ylab
Label for y-axis.
legend
Flag if legend is plotted. If legend == "outer" the legend is plotted in the outer margins of the figure. This is useful if both diagonals are used.
colspace
Either "hcl" or "rgb". Colour space to be used for the plots.
col
If colspace == "hcl", the vector is giving the minimum and maximum values of hue (element 1 & 2), chroma (element 3 & 4) and luminance (element 5 & 6). The optional element 7 is used as alpha value. See hcl for further explanation. If colspace == "rgb", a vector of length >=2 giving the colours to be interpolated using colorRamp.
digits
Precision of labels in legend.
range
"auto" or a vector of length = 2 giving the range of values to be plotted.
constraint
A character string giving a constraint which values should be plotted. See examples section.
uppertriang
Flag if upper triangle is used for the plot. Note that if TRUE the current plot is used instead of starting a new plot
...
Further arguments passed to plot.default.

Value

An invisible vector with minimum and maximum values plotted.

Details

See details in glm.nri and glm.

See Also

nri, glm.nri, glm, cor.test, t.test

Examples

Run this code
## Not run: 
# data(spectral_data)
# 
# ## Calculate all possible combinations for WorldView-2-8
# spec_WV <- spectralResampling(spectral_data, "WorldView2-8",
#                               response_function = FALSE)
# nri_WV <- nri(spec_WV, recursive = TRUE)
# 
# ## Fit generalised linear models between NRI-values and chlorophyll
# glmnri <- glm.nri(nri_WV ~ chlorophyll, preddata = spec_WV)
# 
# ## Plot p-values
# plot(glmnri, range = c(0, 0.05))
# ## Plot t-values
# plot(glmnri, coefficient = "t.value")
# ## Plot only t-values where p-values < 0.001
# plot(glmnri, coefficient = "t.value", 
#      constraint = "p.value < 0.001")
# 
# ## Fit linear models between NRI-values and chlorophyll
# lmnri <- lm.nri(nri_WV ~ chlorophyll, preddata = spec_WV)
# 
# ## Plot r.squared
# plot(lmnri)
# 
# ## Example for EnMAP (Attention: Calculation time may be long!)
# spec_EM <- spectralResampling(spectral_data, "EnMAP", 
#                               response_function = FALSE)
# mask(spec_EM) <- c(300, 550, 800, 2500)
# nri_EM <- nri(spec_EM, recursive = TRUE)
# glmnri <- glm.nri(nri_EM ~ chlorophyll, preddata = spec_EM)
# 
# ## Plot T values in lower and p-values in upper diagonal
# ## of the plot
# ## Enlarge margins for legends
# par(mar = c(5.1, 4.1, 4.1, 5))
# plot(glmnri, coefficient = "t.value", legend = "outer")
# plot(glmnri, coefficient = "p.value", uppertriang = TRUE)
# lines(c(400,1705),c(400,1705))
# ## End(Not run)

Run the code above in your browser using DataLab