scripts/old/test.update.tau.kappa.R

In NorskRegnesentral/SpatGEVBMA: Hierarchical spatial generalized extreme value (GEV) modeling with Bayesian Model Averaging (BMA)

rm(list = ls())

# Run the following line after writing new cpp-code to test.
#Rcpp::compileAttributes("/nr/samba/user/jullum/Prosjekter/FlomQ/Git_SpatialGEVBMA/SpatialGEVBMA")

# Run the following in the terminal:
# cd /nr/samba/user/jullum/Prosjekter/FlomQ/Git_SpatialGEVBMA
# R CMD INSTALL --library="/nr/samba/user/jullum/Rlibs/new_packages" SpatialGEVBMA



library("SpatialGEVBMA", lib.loc="/nr/samba/user/jullum/Rlibs/new_packages")


## Unload
data(norway)
attach(norway)

set.seed(1)
G <- gev.init(Y.list, X, S, NULL, FALSE, NULL,nonspatial=FALSE,log.kappa=FALSE)
for(i in 1:5)G <- gev.update(G)


s=3
n.s <- G$n.s
theta.kappa <- G$theta.kappa

C <- 1/G$alpha.kappa * exp(-1/G$lambda.kappa * G$D)
diag(C) <- diag(C) + 1e-5
C.inv <- solve(C)
tau <- G$tau.kappa



  Y.s <- G$Y.list[[s]]
  mu.s <- sum(G$theta.mu * G$X[s,]) + G$tau.mu[s]
  eps.s <- Y.s - mu.s
  kappa.hat <- sum(G$theta.kappa * G$X[s,])
  xi.s <- sum(G$theta.xi * G$X[s,]) + G$tau.xi[s]
  
  tau.hat.s <- sum(-C.inv[s,-s]/C.inv[s,s] * tau[-s])
  varsigma.s <- 1/C.inv[s,s] ##precision matrix stuff
  

### Testing:

g.s <- g.prime(tau[s], tau.hat.s, varsigma.s, xi.s, kappa.hat, eps.s)
gg.s <- g.double.prime(tau[s], tau.hat.s, varsigma.s, xi.s, kappa.hat, eps.s)

g.s.arma <- g_prime_new(tau[s], tau.hat.s, varsigma.s, xi.s, kappa.hat, eps.s)
gg.s.arma <- g_double_prime_new(tau[s], tau.hat.s, varsigma.s, xi.s, kappa.hat, eps.s)

g.s
g.s.arma

gg.s
gg.s.arma

g.s <- g.prime(tau[s], tau.hat.s, varsigma.s, 0, kappa.hat, eps.s)
gg.s <- g.double.prime(tau[s], tau.hat.s, varsigma.s, 0, kappa.hat, eps.s)

g.s.arma <- g_prime_new(tau[s], tau.hat.s, varsigma.s, 0, kappa.hat, eps.s)
gg.s.arma <- g_double_prime_new(tau[s], tau.hat.s, varsigma.s, 0, kappa.hat, eps.s)

g.s
g.s.arma

gg.s
gg.s.arma



### Identical



ptm=proc.time()
for (i in 1:10^5)
{
  g.s <- g.prime(tau[s], tau.hat.s, varsigma.s, xi.s, kappa.hat, eps.s)
}
proc.time()-ptm
#user  system elapsed 
#2.488   0.000   2.493 

ptm=proc.time()
for (i in 1:10^5)
{
  g.s.arma <- g_prime_new(tau[s], tau.hat.s, varsigma.s, xi.s, kappa.hat, eps.s)
}
proc.time()-ptm
#user  system elapsed 
#1.284   0.008   1.293 

# 50% speedup

ptm=proc.time()
for (i in 1:10^5)
{
  gg.s <- g.double.prime(tau[s], tau.hat.s, varsigma.s, xi.s, kappa.hat, eps.s)
}
proc.time()-ptm
#user  system elapsed 
#2.060   0.000   2.065 

ptm=proc.time()
for (i in 1:10^5)
{
  gg.s.arma <- g_double_prime_new(tau[s], tau.hat.s, varsigma.s, xi.s, kappa.hat, eps.s)
}
proc.time()-ptm
#user  system elapsed 
#1.788   0.004   1.797 

## 10% Speedup