Question

Please bear with me if this is rather tenuous, and feel free to ask questions if I have left anything out…

I’m attempting to do some 50 year extreme wind calculations based on the following link

http://www.wasp.dk/Products/weng/ExtremeWinds.htm

They seem to use the gumbel distribution, so I have used function gumbel in package “evir” to fit the distribution to the data, and function dgumbel in package “evd” as the plotting function.

package("evd")
package("evir")

speeds2 <- data.frame(speed=sample(10:50,1000,rep=TRUE))
gumbel(speeds2$speed)

I have then tried to plot this using ggplot2’s stat_function, like so (except for now I have put in dummy values for loc and scale.

library(ggplot2)
ggplot(data=speeds2, aes(x=speed)) + 
  stat_function(fun=dgumbel, args=list(loc=1, scale=0.5))

I get the following error:

Error in dgev(x, loc = loc, scale = scale, shape = 0, log = log) : 
  unused argument(s) (loc = loc, scale = scale, shape = 0, log = log)

I am unsure if I am doing this the right way. Any pointers would be much appreciated.

Answer 1

Earlier session showed that the parameter estimates from the gumbel call were near 24 and 11.

library(evd)
library(ggplot2)
 speeds2 <- data.frame(speed=sample(10:50,1000,rep=TRUE))
 ggplot(data=speeds2, aes(x=speed), geom="density") + 
   stat_function(fun=dgumbel, args=list(loc=24, scale=11))

If you only used the parameters of 1 and 0.5, you got a straight flat line. Loading only evd prevents conflicts with the dgumbel-related functions in evir. When you load evir second you get:

> speeds2 <- data.frame(speed=sample(10:50,1000,rep=TRUE))
> ggplot(data=speeds2, aes(x=speed), geom="density") + 
+   stat_function(fun=dgumbel, args=list(loc=24, scale=11))
Error in dgev(x, loc = loc, scale = scale, shape = 0, log = log) : 
  unused argument(s) (loc = loc, scale = scale, shape = 0, log = log)

Demonstrating how to make a call to a dgumbel function in a particular (better behaved) package:

library(VGAM)
ggplot(data = speeds2, aes(x = speed)) + 
   stat_function(fun = VGAM::dgumbel, args = list(location = 24, scale = 11))

I think Ramnath’s suggestion to add the empiric ‘density’ is good but I prefer to use geom_histogram:

ggplot(data=speeds2, aes(x=speed)) + geom_histogram(aes(y = ..density..) , binwidth=5 ) + 
                            stat_function(fun=dgumbel, args=list(loc=24, scale=11))