A C++ library for Bayesian modeling, mainly through Markov chain Monte Carlo, but with a few other methods supported. BOOM = "Bayesian Object Oriented Modeling". It is also the sound your computer makes when it crashes.
GNU Lesser General Public License v2.1
35
stars
14
forks
source link
R crashes when trying bsts.prediction.errors for the model with family="student" #42
I found that when trying to compute prediction errors with cutpoints for the model of “Student”-family, R session terminates regardless if it’s R IDE, RStudio, Jupyter Notebook and also either on Mac or Linux.
Here’s the code, based on the sample code from “bsts.prediction.errors” docs:
y <- log(AirPassengers)
ss <- AddLocalLinearTrend(list(), y)
ss <- AddSeasonal(ss, y, nseasons = 12)
model <- bsts(y, state.specification = ss, niter = 500, family="student")
I found that when trying to compute prediction errors with cutpoints for the model of “Student”-family, R session terminates regardless if it’s R IDE, RStudio, Jupyter Notebook and also either on Mac or Linux.
Here’s the code, based on the sample code from “bsts.prediction.errors” docs:
y <- log(AirPassengers)
ss <- AddLocalLinearTrend(list(), y)
ss <- AddSeasonal(ss, y, nseasons = 12)
model <- bsts(y, state.specification = ss, niter = 500, family="student")
errors <- bsts.prediction.errors(model, cutpoints = c(80, 120))