Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextComplex models in ecology: challenges and solutionsPéter Sól...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextComplex models are everywhereˆ Ecology is the scientic study...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextHierarchical modelsˆ Inference:ˆ (y | X = x) ∼ h(y ; X = x, ...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextThe Bayesian toolkitˆ MCMC is easy because others did the he...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextNormalNormal modelˆ Yij | µij ∼ Normal(µij, σ2)ˆ i = 1, . . ...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextNormalNormal model library(rjags) library(dclone) set.seed(1...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextJAGS dat - list(Y = Y, X = X, nm = nm, n = n, np = ncol(X), ...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextJAGS summary(m)Iterations = 3001:8000Thinning interval = 1Nu...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextJAGSsigma0.15 0.40 −1.15 −0.950.550.150.402040tau20 10theta[...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextData cloning (DC)ˆ Basic results1:ˆ y(K)= (y , . . . , y )ˆ ...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextIterative model tting str(dclone(dat, n.clones = 2, unchange...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextMarginal posterior summariessigmaNumber of clones1 2 4 80.54...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextDC diagnostics dcdiag(mk)n.clones lambda.max ms.error r.squa...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextComputational demands1.0 2.0 3.0 4.004812Number of chainsPro...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextHPC to the rescue!ˆ MCMC is labelled as embarassingly parall...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextWorkload optimizationˆ Size balancing for DC:ˆ start with la...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextSize balancing0 2 4 6 8 1221No BalancingMax = 12Approximate ...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextTypes of parallelism ## snow type clusters cl - makeCluster(...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextRandom Number Generationˆ WinBUGS/OpenBUGS: seeds approach d...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextHMC in STANˆ Another way to cut back on burn-in is to use Ha...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextThe two mind-sets1. Analytic mid-setˆ use a predened general...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone Nextsharx: an examplesharx is a package to t hierarchical specie...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone Nextdcmleˆ The dcmle package was motivated by stats4:::mle and t...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone Nextseeds example sourceDcExample(seeds) seedsFormal class dcFit...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone Nextseeds example custommodel(seeds@model)Object of class custom...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone Nextseeds example dcm - dcmle(seeds, n.clones = 1:3, n.iter = 10...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextPVACloneˆ Likelihood based population viability analysis in ...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextGrowth model objects library(PVAClone) gm - ricker(normal, f...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextModel with xed parameters gm@modelObject of class custommode...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextLikelihood ratio test (DCLR) m1 - pva(redstart, gompertz(nor...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextProle likelihood alt - pva(redstart, ricker(normal, fixed = ...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextProle likelihood−0.10 −0.05 0.00 0.05 0.10 0.15 0.20−300−200...
Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextWhats next?ˆ Things done:ˆ DC support for OpenBUGS, WinBUGS,...
Upcoming SlideShare
Loading in …5
×

Complex models in ecology: challenges and solutions

862 views

Published on

41st Annual Meeting of the SSC -- Recent developments in R packages, May 27, 2013

Published in: Technology, Spiritual
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
862
On SlideShare
0
From Embeds
0
Number of Embeds
66
Actions
Shares
0
Downloads
13
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

Complex models in ecology: challenges and solutions

  1. 1. Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextComplex models in ecology: challenges and solutionsPéter Sólymoswith K. Nadeem and S. R. LeleUniversity of Alberta41st Annual Meeting of the SSCRecent developments in R packagesMay 27, 2013
  2. 2. Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextComplex models are everywhereˆ Ecology is the scientic study of interactions of organisms with one another andwith their environment.ˆ Data is growing fast, models are becoming more complex.ˆ We need complex models for dealing with:ˆ non-independence (spatial, temporal, phylogenetic),ˆ missing data,ˆ observation and measurement error.
  3. 3. Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextHierarchical modelsˆ Inference:ˆ (y | X = x) ∼ h(y ; X = x, θ1)ˆ X ∼ g (x; θ2)ˆ θ = (θ1, θ2)ˆ L(θ; y ) = h(y | x; θ1)g (x; θ2)dxˆ Computation:ˆ high dimensional integral hard to calculate,ˆ noisy likelihood surface numerical search is hard,ˆ second derivatives hard to calculate.ˆ life is hard if you are a frequentist.
  4. 4. Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextThe Bayesian toolkitˆ MCMC is easy because others did the heavy lifting:ˆ WinBUGS, OpenBUGSˆ JAGSˆ STANˆ Great interfaces with R:ˆ R2WinBUGS, R2OpenBUGS, BRugsˆ coda, rjagsˆ rstanˆ Inference based on the posterior distribution:ˆ π(θ | y ) =L(θ;y)π(θ)L(θ;y)π(θ)dθ,ˆ π(θ) is the prior distribution.
  5. 5. Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextNormalNormal modelˆ Yij | µij ∼ Normal(µij, σ2)ˆ i = 1, . . . , n; j = 1, . . . , mnˆ µij = XTij θ + iˆ i ∼ Normal(0, τ2)model {for (ij in 1:nm) { #### - likelihoodY[ij] ~ dnorm(mu[ij], 1/sigma^2)mu[ij] - inprod(X[ij,], theta) + e[gr[ij]]}for (i in 1:n) {e[i] ~ dnorm(0, 1/tau^2)}for (k in 1:np) { #### - priorstheta[k] ~ dnorm(0, 0.001)}sigma ~ dlnorm(0, 0.001)tau ~ dlnorm(0, 0.001)}
  6. 6. Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextNormalNormal model library(rjags) library(dclone) set.seed(1234) theta - c(1, -1) sigma - 0.6 tau - 0.3 n - 50 # number of clusters m - 10 # number of repeats within each cluster nm - n * m # total number of observations gr - rep(1:n, each = m) # group membership defining clusters x - rnorm(nm) # covariate X - model.matrix(~x) # design matrix e - rnorm(n, 0, tau) # random effect mu - drop(X %*% theta) + e[gr] # mean Y - rnorm(nm, mu, sigma) # outcome
  7. 7. Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextJAGS dat - list(Y = Y, X = X, nm = nm, n = n, np = ncol(X), gr = gr) str(dat)List of 6$ Y : num [1:500] 1.669 0.34 -0.474 3.228 0.968 ...$ X : num [1:500, 1:2] 1 1 1 1 1 1 1 1 1 1 .....- attr(*, dimnames)=List of 2.. ..$ : chr [1:500] 1 2 3 4 ..... ..$ : chr [1:2] (Intercept) x..- attr(*, assign)= int [1:2] 0 1$ nm: num 500$ n : num 50$ np: int 2$ gr: int [1:500] 1 1 1 1 1 1 1 1 1 1 ... m - jags.fit(dat, c(theta, sigma, tau), model, n.update = 2000)
  8. 8. Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextJAGS summary(m)Iterations = 3001:8000Thinning interval = 1Number of chains = 3Sample size per chain = 50001. Empirical mean and standard deviation for each variable,plus standard error of the mean:Mean SD Naive SE Time-series SEsigma 0.581 0.0193 0.000158 0.000225tau 0.279 0.0410 0.000335 0.000739theta[1] 0.959 0.0479 0.000391 0.000959theta[2] -1.032 0.0260 0.000213 0.0002302. Quantiles for each variable:2.5% 25% 50% 75% 97.5%sigma 0.546 0.568 0.581 0.594 0.621tau 0.205 0.250 0.276 0.305 0.365theta[1] 0.862 0.927 0.959 0.991 1.052theta[2] -1.083 -1.050 -1.032 -1.014 -0.981
  9. 9. Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextJAGSsigma0.15 0.40 −1.15 −0.950.550.150.402040tau20 10theta[1]0.81.10.55−1.15−0.9550200.8 1.110theta[2]
  10. 10. Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextData cloning (DC)ˆ Basic results1:ˆ y(K)= (y , . . . , y )ˆ L(θ; yK) = L(θ; y )Kˆ πK(θ | y ) =[L(θ;y)]Kπ(θ)[L(θ;y)]Kπ(θ)dθ,ˆ πK(θ | y ) ∼ MVN(ˆθ, 1K I−1(ˆθ))ˆ Implications:ˆ we can use Bayesian MCMC toolkit for frequentist inference,ˆ mean of the posterior is the MLE (ˆθ),ˆ K times the posterior variance is the variance of the MLE.ˆ High dimensional integral no need to calculate,ˆ noisy likelihood surface no numerical optimization involved,ˆ second derivatives no need to calculate.ˆ This is independent of the specication of the prior distribution.1Lele et al. 2007 ELE; Lele et al. 2010 JASA
  11. 11. Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextIterative model tting str(dclone(dat, n.clones = 2, unchanged = np, multiply = c(nm,+ n)))List of 6$ Y : atomic [1:1000] 1.669 0.34 -0.474 3.228 0.968 .....- attr(*, n.clones)= atomic [1:1] 2.. ..- attr(*, method)= chr rep$ X : num [1:1000, 1:2] 1 1 1 1 1 1 1 1 1 1 .....- attr(*, dimnames)=List of 2.. ..$ : chr [1:1000] 1_1 2_1 3_1 4_1 ..... ..$ : chr [1:2] (Intercept) x..- attr(*, n.clones)= atomic [1:1] 2.. ..- attr(*, method)= chr rep$ nm: atomic [1:1] 1000..- attr(*, n.clones)= atomic [1:1] 2.. ..- attr(*, method)= chr multi$ n : atomic [1:1] 100..- attr(*, n.clones)= atomic [1:1] 2.. ..- attr(*, method)= chr multi$ np: int 2$ gr: atomic [1:1000] 1 1 1 1 1 1 1 1 1 1 .....- attr(*, n.clones)= atomic [1:1] 2.. ..- attr(*, method)= chr rep mk - dc.fit(dat, c(theta, sigma, tau), model, n.update = 2000,+ n.clones = c(1, 2, 4, 8), unchanged = np, multiply = c(nm,+ n))
  12. 12. Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextMarginal posterior summariessigmaNumber of clones1 2 4 80.540.60x x x xxxxxtauNumber of clonesEstimate1 2 4 80.200.35xx x xxx x xtheta[1]1 2 4 80.901.00x x x xx x x xtheta[2]Estimate1 2 4 8−1.08−1.00xxx xxxx x
  13. 13. Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextDC diagnostics dcdiag(mk)n.clones lambda.max ms.error r.squared r.hat1 1 0.002150 0.04172 0.003217 1.0012 2 0.002234 0.08290 0.006286 1.0083 4 0.002322 0.10398 0.010156 1.0014 8 0.002117 0.08042 0.006313 1.012ˆ It can help in dientifying the number of clones required.ˆ Non-identiability can be spotted as a bonus.
  14. 14. Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextComputational demands1.0 2.0 3.0 4.004812Number of chainsProcessingtime(sec)1 2 3 4 5 6 7 80204060Number of clonesProcessingtime(sec)
  15. 15. Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextHPC to the rescue!ˆ MCMC is labelled as embarassingly parallel problem.ˆ Distributing independent chanins to workers:ˆ Proper initialization (initial values, RNGs),ˆ run the chains on workers,ˆ collect results (large MCMC object might mean more communication overhead),ˆ repeat this for dierent number of clones.ˆ The paradox of burn-in:ˆ one long chain: burn-in happens only once,ˆ few-to-many chains: best trade-o w.r.t. burn-in,ˆ n.iter chains: burn-in happens n.iter times (even with ∞ chains).ˆ Computing time should drop to 1/n.iter + overhead.ˆ This works for Bayesian analysis and DC.ˆ Learning can happen, i.e. results can be used to make priors more informative(this improves mixing and reduces burn-in).
  16. 16. Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextWorkload optimizationˆ Size balancing for DC:ˆ start with largest problem chighest K ,ˆ run smaller problems on other workers.ˆ collect results (collect whole MCMC object for highest K , only summaries forothers),ˆ Learning is not an option here, need to have good guesses or rely onnon-informative priors.ˆ Can be combined with the parallel chains approach.ˆ Computing time is dened by the highest K problem.
  17. 17. Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextSize balancing0 2 4 6 8 1221No BalancingMax = 12Approximate Processing TimeWorkers1 23 4 50 2 4 6 8 1221Size BalancingMax = 8Approximate Processing TimeWorkers5 2 14 3
  18. 18. Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextTypes of parallelism ## snow type clusters cl - makeCluster(3) m - jags.parfit(cl, ...) mk - dc.parfit(cl, ...) stopCluster(cl) ## muticore type forking m - jags.parfit(3, ...) mk - dc.parfit(3, ...)ˆ Parallel chains approach not available for WinBUGS/OpenBUGS,ˆ dc.parfit allows size balancing for WinBUGS/OpenBUGS.ˆ Forking does not work on Windows.ˆ (STAN: all works, see R-Forge2)2http://dcr.r-forge.r-project.org/extras/stan.t.R
  19. 19. Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextRandom Number Generationˆ WinBUGS/OpenBUGS: seeds approach does not guarantee independence of verylong chains.ˆ STAN uses LEcuyer RNGs.ˆ JAGS: uses 4 RNGs in base module, but there is the lecuyer module whichallows high number of independent chains. ## base::BaseRNG factory str(parallel.inits(NULL, 2))List of 2$ :List of 2..$ .RNG.name : chr base::Marsaglia-Multicarry..$ .RNG.state: int [1:2] 2087563718 113920118$ :List of 2..$ .RNG.name : chr base::Super-Duper..$ .RNG.state: int [1:2] 1968324500 1720729645 ## lecuyer::RngStream factory load.module(lecuyer, quiet = TRUE) str(parallel.inits(NULL, 2))List of 2$ :List of 2..$ .RNG.name : chr lecuyer::RngStream..$ .RNG.state: int [1:6] -1896643356 145063650 -1913397488 -341376786 297806844 1434416058$ :List of 2..$ .RNG.name : chr lecuyer::RngStream..$ .RNG.state: int [1:6] -1118475959 1854133089 -159660578 -80247816 -567553258 -1472234812 unload.module(lecuyer, quiet = TRUE)
  20. 20. Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextHMC in STANˆ Another way to cut back on burn-in is to use Hamiltonian Monte Carlo andrelated algorithms (No-U-Turn sampler, NUTS).ˆ This is also helpful when parameters are correlated.ˆ See http://mc-stan.org/.ˆ DC support exists3: currently not through CRAN because rstan is not hosted onCRAN (might never be).3http://dcr.r-forge.r-project.org/extras/stan.t.R
  21. 21. Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextThe two mind-sets1. Analytic mid-setˆ use a predened general model,ˆ possibly t it to many similar data sets,ˆ not that interested in algorithms (...),ˆ want something like this: FIT - MODEL(y ~ x, DATA, ...)2. Algorithmic mid-setˆ t a specic modelˆ to a specic data set,ˆ more focus on algorithmic settings: DATA - list(y = y, x = x) MODEL - y ~ x FIT - WRAPPER(DATA, MODEL, ...)How do we provide estimating procedures for folks with an analytic mind-set?
  22. 22. Motivation Bayesian tools Data cloning HPC dcmle PVAClone Nextsharx: an examplesharx is a package to t hierarchical speciesare relationship models (a kind ofmultivariate mixed model for mata-analysis). library(sharx) hsarxfunction (formula, data, n.clones, cl = NULL, ...){if (missing(n.clones))stop(n.clones argument missing)if (missing(data))data - parent.frame()tmp - parse_hsarx(formula, data)dcf - make_hsarx(tmp$Y, tmp$X, tmp$Z, tmp$G)dcm - dcmle(dcf, n.clones = n.clones, cl = cl, nobs = length(tmp$Y),...)out - as(dcm, hsarx)title - if (ncol(tmp$X) 2)SARXelse SARif (!is.null(tmp$Z)) {if (title != SARX NCOL(tmp$Z) 1)title - paste(title, X, sep = )title - paste(H, title, sep = )}out@title - paste(title, Model)out@data - dcfout}environment: namespace:sharx
  23. 23. Motivation Bayesian tools Data cloning HPC dcmle PVAClone Nextdcmleˆ The dcmle package was motivated by stats4:::mle and the modeltoolspackage.ˆ Wanted to provide:ˆ a wrapper around wrappers around wrappers (another abstraction layer),ˆ unied S4 object classes for data and tted models for Bayesian analysis and DC,ˆ lots of methods for access, coercion, summaries, plots.ˆ This is the engine for package development with DC.ˆ Classic BUGS examples:module glm loaded library(dcmle) as.character(listDcExamples()$topic)[1] blocker bones dyes epil[5] equiv leuk litters lsat[9] mice oxford pump rats[13] salm seeds air alli[17] asia beetles biops birats[21] cervix dugongs eyes hearts[25] ice jaw orange pigs[29] schools paramecium
  24. 24. Motivation Bayesian tools Data cloning HPC dcmle PVAClone Nextseeds example sourceDcExample(seeds) seedsFormal class dcFit [package dcmle] with 10 slots..@ multiply : chr N..@ unchanged: NULL..@ update : NULL..@ updatefun: NULL..@ initsfun : NULL..@ flavour : chr jags..@ data :List of 5.. ..$ N : num 21.. ..$ r : num [1:21] 10 23 23 26 17 5 53 55 32 46 ..... ..$ n : num [1:21] 39 62 81 51 39 6 74 72 51 79 ..... ..$ x1: num [1:21] 0 0 0 0 0 0 0 0 0 0 ..... ..$ x2: num [1:21] 0 0 0 0 0 1 1 1 1 1 .....@ model :function ()..@ params : chr [1:5] alpha0 alpha1 alpha2 alpha12 .....@ inits :List of 5.. ..$ tau : num 1.. ..$ alpha0 : num 0.. ..$ alpha1 : num 0.. ..$ alpha2 : num 0.. ..$ alpha12: num 0
  25. 25. Motivation Bayesian tools Data cloning HPC dcmle PVAClone Nextseeds example custommodel(seeds@model)Object of class custommodel:model{alpha0 ~ dnorm(0.00000E+00, 1.00000E-06)alpha1 ~ dnorm(0.00000E+00, 1.00000E-06)alpha2 ~ dnorm(0.00000E+00, 1.00000E-06)alpha12 ~ dnorm(0.00000E+00, 1.00000E-06)tau ~ dgamma(0.001, 0.001)sigma - 1/sqrt(tau)for (i in 1:N) {b[i] ~ dnorm(0.00000E+00, tau)logit(p[i]) - alpha0 + alpha1 * x1[i] + alpha2 * x2[i] +alpha12 * x1[i] * x2[i] + b[i]r[i] ~ dbin(p[i], n[i])}}
  26. 26. Motivation Bayesian tools Data cloning HPC dcmle PVAClone Nextseeds example dcm - dcmle(seeds, n.clones = 1:3, n.iter = 1000) summary(dcm)Maximum likelihood estimation with data cloningCall:dcmle(x = seeds, n.clones = 1:3, n.iter = 1000)Settings:start end thin n.iter n.chains n.clones1001 2000 1 1000 3 3Coefficients:Estimate Std. Error z value Pr(|z|)alpha0 -0.5556 0.1738 -3.20 0.0014 **alpha1 0.0981 0.2909 0.34 0.7360alpha12 -0.8319 0.3947 -2.11 0.0350 *alpha2 1.3547 0.2542 5.33 9.8e-08 ***sigma 0.2450 0.1244 1.97 0.0490 *---Signif. codes: 0 *** 0.001 ** 0.01 * 0.05 . 0.1 1Convergence:n.clones lambda.max ms.error r.squared r.hat1 0.494 NA NA 1.052 0.184 NA NA 1.043 0.130 NA NA 1.02
  27. 27. Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextPVACloneˆ Likelihood based population viability analysis in the presence of observation errorand missing data4.ˆ Builds upon JAGS, dclone, and dcmle.ˆ Growth models: Xt | σ2 ∼ Normal(m(Xt−1, η), σ2)ˆ gompertz: m(Xt−1) = Xt−1 + a + bXt−1,ˆ ricker: m(Xt−1) = Xt−1 + a + beXt−1 ,ˆ bevertonholt,ˆ thetalogistic,ˆ thetalogistic_D.ˆ Observation error: Yt ∼ f (yt; Xt, Ψ)ˆ none,ˆ poisson,ˆ normal,4Nadeem and Lele 2012 OIKOS
  28. 28. Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextGrowth model objects library(PVAClone) gm - ricker(normal, fixed = c(sigma = 0.5)) str(gm)Formal class pvamodel [package PVAClone] with 14 slots..@ growth.model: chr ricker..@ obs.error : chr normal..@ model :Class custommodel chr [1:20] model { for (i in 1:kk) { N[1,i] -..@ genmodel :function ()..@ p : int 4..@ support : num [1:4, 1:2] -Inf -Inf 2.22e-16 2.22e-16 Inf ..... ..- attr(*, dimnames)=List of 2.. .. ..$ : chr [1:4] a b sigma tau.. .. ..$ : chr [1:2] Min Max..@ params : chr [1:3] a b lntau..@ varnames : chr [1:4] a b sigma tau..@ fixed : Named num 0.5.. ..- attr(*, names)= chr sigma..@ fancy : chr [1:2] Ricker Normal..@ transf :function (mcmc, obs.error)..@ backtransf :function (mcmc, obs.error)..@ logdensity :function (logx, mle, data, null_obserror = FALSE, alt_obserror = FALSE)..@ neffective :function (obs)
  29. 29. Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextModel with xed parameters gm@modelObject of class custommodel:model {for (i in 1:kk) {N[1,i] - exp(y[1,i])x[1,i] - y[1,i]for (j in 2:T) {x[j,i] ~ dnorm(mu[j,i], prcx)mu[j,i] - a + b * N[j-1,i] + x[j-1,i]N[j,i] - min(exp(x[j,i]), 10000)y[j,i] ~ dnorm(x[j,i], prcy)}}sigma - 0.5tau - exp(lntau)lnsigma - log(sigma)lntau ~ dnorm(0, 1)a ~ dnorm(0, 0.01)b ~ dnorm(0, 10)prcx - 1/sigma^2prcy - 1/tau^2}
  30. 30. Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextLikelihood ratio test (DCLR) m1 - pva(redstart, gompertz(normal), 50, n.update = 2000,+ n.iter = 1000) m2 - pva(redstart, ricker(normal), 50, n.update = 2000, n.iter = 1000) ms - model.select(m1, m2) coef(m2)a b sigma tau0.07159 -0.01721 0.05096 0.58996 msPVA Model Selection:Time series with 30 observations (missing: 0)Null Model: m1Gompertz growth model with Normal observation errorAlternative Model: m2Ricker growth model with Normal observation errorlog_LR delta_AIC delta_BIC delta_AICc1 -249.6 499.3 499.4 499.3Alternative Model is strongly supported over the Null Model
  31. 31. Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextProle likelihood alt - pva(redstart, ricker(normal, fixed = c(sigma = 0.05)),+ 50, n.update = 2000, n.iter = 1000) p - generateLatent(alt, n.chains = 1, n.iter = 10000) a - c(-0.1, -0.05, 0, 0.05, 0.1, 0.15, 0.2) llr_res - numeric(length(a)) for (i in seq_len(length(a))) {+ null - pva(redstart, ricker(normal, fixed = c(a = a[i],+ sigma = 0.05)), 50, n.update = 2000, n.iter = 1000)+ llr_res[i] - pva.llr(null, alt, pred = p)+ }
  32. 32. Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextProle likelihood−0.10 −0.05 0.00 0.05 0.10 0.15 0.20−300−200−1000aProfilelog−likelihood
  33. 33. Motivation Bayesian tools Data cloning HPC dcmle PVAClone NextWhats next?ˆ Things done:ˆ DC support for OpenBUGS, WinBUGS, JAGS, STAN.ˆ Support for parallel computing.ˆ dcmle engine for package development (sharx, PVAClone, and soon detect).ˆ Things to do:ˆ Full integration with STAN (dc.fit, dcmle).ˆ More examples.ˆ Prediction/forecasting features for PVAClone.ˆ Find out more:ˆ Sólymos 2010 R Journal [PDF]ˆ http://dcr.r-forge.r-project.org/

×