Dispersal Kernel might be user-supplied?

[Lobz]

Related to this, maybe the kernel might be user-supplied? For instance, exponential kernel instead of Gaussian; or gamma?

Option one, we could allow options like "gamma", "normal"... Although we would have to allow more than one parameter for gamma?

Or we could actually allow the user to supply the kernel as a function o.o something like

facByRates((...), dispersionKernel = function(x){rnorm(1,0,x)}, dispersionParameters = list(x=2))

I do believe that this is doable in Rcpp, although, maybe using only pre-existing functions might be easier to implement, something like this:

facByRates((...), dispersionKernel = rnorm, dispersionParameters = list(mean=0,sd=2))

[andrechalom]

I think pre-existing functions really cover the relevant use cases, which I
think would be exponential, normal and gamma (to model some dispersion that
is not concentrated around the original tree)

On Sun, May 22, 2016 at 3:13 PM, M Salles notifications@github.com wrote:

Related to this, maybe the kernel might be user-supplied? For instance,
exponential kernel instead of Gaussian; or gamma?

Option one, we could allow options like "gamma", "normal"... Although we
would have to allow more than one parameter for gamma?

Or we could actually allow the user to supply the kernel as a function o.o
something like

facByRates((...), dispersionKernel = function(x){rnorm(1,0,x)},
dispersionParameters = list(x=2))

I do believe that this is doable in Rcpp, although, maybe using only
pre-existing functions might be easier to implement, something like this:

facByRates((...), dispersionKernel = rnorm, dispersionParameters =
list(mean=0,sd=2))

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub
#20