missing data
think a lot about the interface: how to optimize efficiency, flexibility, breadth, transparency, simplicity ... ? Want simple model objects (S3 class) that allow simple summaries, diagnostics, fits ... extend mle2
class?
mle2
class be extended via S4? or export coef
?summary.pars
gets Roxygenized documentation confused)erlangSEIR
stuff?pomp
, JAGS
, etc. ?EpiModel
, EpiDynamics
, pomp
...check Hessians (analytical + finite-diff of grad, which you can get via numDeriv)
do this for real epidemic curves
compare:
visualization
try with 2 parameters (fix gamma, N, show likelihood surface for beta, I0); 3 parameters?
reparameterization for most independent results?
understand why NLL vs least-squares should matter.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.