AIC and other IT indexes criteria for for backward, forward and stepwise regression

  This is a reasonable question, but it isn't at all specific to mixed
models (which is the topic of this mailing list).  You could try
CrossValidated (https://stats.stackexchange.com).

  I'm sure opinions differ a lot, and answers will almost certainly
depend on your goals and context, but *if* I were going to do model
selection (which I think is very often a bad idea!) I would simply pick
the model with the minimum AIC, which will (asymptotically) have the
smallest expected Kullback-Leibler distance.

On 2019-12-18 6:07 a.m., Mario Garrido wrote:

Dear users,
Im currently exploring on the use of AIC and other I-T indexes criteria

for

backward, forward and stepwise regression.
Usually, when applying IT indexes for Multimodal Inference, we choose a

set

of 'good models' depending on different criteria, but mainly, all models
with delta AIC<2, and then we averaged the estimates between the set of
models or make conclusions based on the set of models, no need to

average.

However, if Im not wrong, the goal of backward etc is to get to one

'best'

final model. I understand the use of AIC in this framework but, is there
any criteria to select the best model in this case? Do I simply have to
choose the model with the lowest AIC no matter whether there is another
model whose delta is less than 2? Does it depend on a personal criteria?
For example, if my 'maximal' or saturated model has the lowest AIC and

the

model dropping one variable has a delta of 0.5, which model to choose?
I was looking on the web and I have found no answer to this. So, any
literature recommendation or advice will be welcome.
Thanks