I have a set of Monte Carlo simulation results for a Quick Response
Freight Model that I am currently preparing summary graphs. I want to
show three things on the graphs; the model forecast, an approximate
exponential best fit line, and a smooth line through mean+sd and mean-sd
data points.
As you can see I can get the first two, but all I can find when
searching the help are references to fitting spline models to the data
points. I'm old school, so I was used to using a french curve to draw
the smooth lines I'm interested in. The problem is that I don't
understand how/why of fitting a model to the data points.
Here is an example of the data I am graphing:
year mean sd
1 2010 6738 1721
2 2015 8793 2326
3 2020 11699 3333
4 2025 15951 5232
5 2030 22492 9768
6 2035 33289 24431
7 2040 52917 82341
And here is the R code I am currently using:
gr = function (yr,rate,trk) trk*rate^(yr-2006)
plot.results <- function(dfn) {
dt <- read.table(dfn)
res.g = nls(dt$mean ~ gr(dt$year, R, T), start=c(R=1.08, T=3700),
data=dt)
R <- as.numeric(coef(res.g)[1])
T <- as.numeric(coef(res.g)[2])
plot(dt$year,dt$mean, pch=1, col="green", main="Truck
Forecast",
xlab="Year", ylab="Truck Volume",
ylim=c(0,max(dt$year+dt$sd)))
curve(gr(x, R, T), add=TRUE, col="green")
}
Any help is appreciated!
Walter Anderson, GISP
PBS&J
6504 Bridge Point Pkwy, Ste. 200
Austin, TX. 78730
P:512.342.3237
F:512.327.2453
M:512.364.8318
[[alternative HTML version deleted]]
