[R] help--kernel distribution dynamics

Eugene Salinas eugenesalinas2003 at yahoo.com
Sat Aug 23 20:02:53 CEST 2003


--- Deepayan Sarkar <deepayan at stat.wisc.edu> wrote:
[...]
> 
> As a first step, you could create a matrix (with 50
> rows, one for each time 
> point) where each row holds the kernel density
> estimate for that time point. 
> e.g. (with a grid of size 100 for each estimated
> density),
> 
> foo <- matrix(0, 50, 100)
> for (i in 1:50) 
>     foo[i, ] <- density(rnorm(5000), from = -4, to =
> 4, n = 100)$y
>                         ^^^^^^^^^^^       
> ^^^^^^^^^^^
>                          put your        put
> appropriate 
>                        variable here          ranges
> 
> Whether a 3D view of this will be very informative
> will depend on your data 
> (maybe you could play with the density()
> parameters), but persp() should give 
> you something:
> 
> persp(foo, theta = 135, phi = 30, scale = FALSE,
>       ltheta = -120, shade = 0.75, border = NA)
> 

Hi, Thanks a lot. This seems like what I want to do. I
don't know all the syntax yet so just a
clarification...

Is the  [....]n = 100)$y there in order to condition
on y which is the year and derive the conditional
kernel density? The structure of my data looks like
this
             x  year
Indiv1 1950  .    .
Indiv1 1951  .    .
...................
Indiv1 1991  .    .
...................
...................
IndivN 1991  .    .

so to get the matrix of densities I would write...

for (i in 1950:1991) 
>     foo[i-1949, ] <- density(x, from = -nn, to =nn,
n = 100)$year

Am I right? thanks, e.




More information about the R-help mailing list