dic---deviance information criterion---is a specific information measure, based on generalizing aic/bic, to be used in model choice. See https://en.wikipedia.org/wiki/Deviance_information_criterion
Questions tagged [dic]
10 questions
3
votes
1 answer
Calculating DIC from MCMC (gibbs) output
I just want to make sure my understanding of how we estimate DIC using MCMC output is correct, as the wikipedia page is somewhat confusing.
Defining DIC by
$$DIC = p_D + \bar{D}$$
with
$$D(\theta) = -2 \log p(y|\theta) + C$$
$$p_D = \bar{D} -…
Xiaomi
- 2,276
- 6
- 19
2
votes
1 answer
DIC, WAIC in JAGS
I have a Bayesian Hierarchical model using JAGS. In order to find the best model, I have compared the DIC of two models but It's not reliable. So, I decided to calculate WAIC from JAGS. However I have a big ambiguity and i would appreciate if you…
skhalighi
- 21
- 2
1
vote
0 answers
Find DIC values for another sample
I trained a model using JAGS and I obtained parameter values. Can I calculate the DIC values for a subset of these data points. The code is
mod=jags.model(textConnection(mod_string), data=data_jags, n.chains=3)
dic= dic.samples(mod,…
A Doe
- 15
- 4
1
vote
0 answers
DIC to compare models with different numbers of parameter?
I am interested in comparing hierarchical Bayesian models based on the same dataset but differing in their spatial and temporal resolution. In short, I am looking at animal population changes over time in a large area. I want to compare models that…
Gigi
- 11
- 1
1
vote
0 answers
DIC model selection: preferred model does not contain significant term
I estimate model parameters for 2 glmm models using MCMC in R2JAGS. I then use DIC to assess the relative fit of the models:
Model 1 y~ site + treatment (DIC = 922.1)
Model 2 y~ site (DIC = 921.8)
Given the low Δ DIC between models 1 and 2, I…
RBW
- 41
- 4
1
vote
0 answers
DIC vs PED in JAGS
JAGS allows reporting of DIC and also PED (penalized expected deviance). What are differences between these two statistics? Which is better for what purpose? DIC was the first one, PED was evidently added later, so what was the reason for…
Tomas
- 5,735
- 11
- 52
- 93
1
vote
1 answer
Using DIC for model selection: (in)valid comparisons
Imagine I have the following 7 glmm models where $b_1$ through $b_3$ are fixed effects.
$M_1 = y \sim b_1 \times b_2 \times b_3$
$M_2 = y \sim b_2 \times b_3$
$M_3 = y \sim b_1 \times b_3$
$M_4 = y \sim b_1 \times b_2$
$M_5 = y \sim b_3$
$M_6 = y…
rg255
- 752
- 2
- 8
- 27
0
votes
0 answers
I'm doing model selection using DIC and WAIC. Can they be used to compare different models?
I'm comparing fitted models using the same dataset.
The Models I am tweaking are Logit, Probit and a variation of Probit(Binary models).
I can use DIC and WAIC to make these comparisons.
Do you have a bibliographic reference that provides arguments…
0
votes
0 answers
Is it okay to have a negative LOOic?
I have a set of 9 models that I am running using R package MixSIAR to investigate the diet of Bottlenose dolphins in an estuary. The models have combinations of factors with random (social cluster, individual) and fixed (age, sex) factors. Two of…
0
votes
0 answers
Deviance Information Criterion
I am trying two compare two hierarchical Bayesian models using the following output from R:
I have understood that the DIC is used to compare Bayesian hierarchical models and that the lowest its value, the better.
Basically, I'm having two models…
jmemax
- 1