MARK book - AICc and Deviance values from aa.inp data

[trrobbin]

Hello,
running through the examples in the book, the swift data from aa.inp specifically, I am finding a discrepancy between the AICc values (and the measures that are based on them), but not the deviance values nor the number of parameters. So, when I run the models, which I think are being run correctly, the deviance values and the no. Parameters are identical to what is in the book, but the AICc values are not. They are about 3 measures off.

I have searched the forum, thinking (hoping) that I am not the only one that had this problem, however, I was unable to find anything. I have run it multiples times from the beginning and keep getting the same results.

What might I be doing wrong?

Thank you!

[cooch]

Hello,
running through the examples in the book, the swift data from aa.inp specifically, I am finding a discrepancy between the AICc values (and the measures that are based on them), but not the deviance values nor the number of parameters. So, when I run the models, which I think are being run correctly, the deviance values and the no. Parameters are identical to what is in the book, but the AICc values are not. They are about 3 measures off.

I have searched the forum, thinking (hoping) that I am not the only one that had this problem, however, I was unable to find anything. I have run it multiples times from the beginning and keep getting the same results.

What might I be doing wrong?

Thank you!

Not sure - but in part thats because when you make this sort of post, you need to specify what chapter, what page, what model etc etc. Specific details will help us track down if/where there might be discrepancies.

[cooch]

Hello,
running through the examples in the book, the swift data from aa.inp specifically, I am finding a discrepancy between the AICc values (and the measures that are based on them), but not the deviance values nor the number of parameters. So, when I run the models, which I think are being run correctly, the deviance values and the no. Parameters are identical to what is in the book, but the AICc values are not. They are about 3 measures off.

Think I found the particular 'results browser' you were referring too, and what the problem was. Not so long ago, I consolidated chapters 4 and 6 into a single chapter (current chapter 4), and it seems as if an old - incorrect figure - got put into the new chapter 4. I've corrected it, and uploaded a new chapter 4.

[trrobbin]

Yes, that's it. Thank you, and I'm sorry for not including the specific placement in the book.

[cooch]

Yes, that's it. Thank you, and I'm sorry for not including the specific placement in the book.

No problem - just wanted to provide you (and everyone else) with the suggested procedure for reporting a bug (software, documentation, or otherwise) - the more detail you can provide, the easier it is to track things down.