HEAT (version 1.2)

plot.threshpt: Plots from a threshpt Object

Description

Following four plots from a fitted threshpt object are provided: Plot 1: plot for a response variable on the natural cubic spline of a exposure variable Plot 2: plot for fitted response values with other mean covariates and 95% confidence intervals Plot 3: plot for fitted response values Plot 4: deviance by threshold point plot

Usage

"plot"(x, select = NULL, se = T, expdf = 4, 
xlim = NULL, ylim = NULL, xaxt = NULL, yaxt = NULL, 
col.value = NULL, col.preval = NULL, col.ci = NULL, col.vline = NULL,
lwd = NULL, pch = NULL, pch.preval = NULL, main = NULL, xlab = NULL, ylab = NULL, ...)

Arguments

x

A fitted threshpt object produced by threshpt()

select

Select a plot type

se

When TRUE (default) upper and lower confidence interval lines are added to the plot 1 and 4. When FALSE, otherwise.

expdf

Degree of freedom of natural cubic spline function for the main exposure variable in plot 1; default value is four.

xlim

The x limits of the plot.

ylim

The y limits of the plot.

xaxt

A character which specifies the x axis type.

yaxt

A character which specifies the y axis type.

col.value

The color of main values or line in all kind of plot.

col.preval

The color of fitted value in plot 3.

col.ci

The color of confidence interval lines in plot 1 and 3.

col.vline

The color of vertical line that represents a optimum threshold in plot 2, 3 and 4.

lwd

Line width for plot 1

pch

Either an integer specifying a symbol or a single character to be used in plotting points in plot 2 and 3.

pch.preval

Either an integer specifying a symbol or a single character to be used in plotting fitted values in plot 3.

main

Overall title of the plot.

xlab

A title for the x axis of the plot

ylab

A title for the y axis of the plot

...

Not used.

Value

Generated four kind of plots from a fitted threshpt object

Examples

# read the Seoul data set and create lag variables
data(mort)
seoul = read6city(mort, 11)
seoul_lag = lagdata(seoul, c("meantemp", "mintemp", "meanpm10", "meanhumi"), 5)

# find a optimal threshold and conduct piecewise linear regression
mythresh = threshpt(nonacc ~ meantemp_m3 + meanpm10_m2 +  meanhumi + ns(sn, 4*10) + factor(dow), 
			     expvar = "meantemp_m3", family = "poisson", data = seoul_lag,
			     startrng = 23, endrng = 33, searchunit = 0.2)

# obtain plots
plot(mythresh, select = 1, se = TRUE, expdf=8, col.value = "blue", col.ci = "light blue")
plot(mythresh, select = 2, se = FALSE, col.vline = "orange")
plot(mythresh, select = 3, pch = 1, pch.preval = 2)
plot(mythresh, select = 4)