Bootstrap Success Rate in Phoenix Software

[crystal]

Dear Forum Members,

I am currently working on a project involving bootstrap analysis, and when using the Nommem, I can find the successful rate of bootstrap simulations. However, I have not been successful in finding the specific location where the success rate of bootstrap simulations is displayed in Phoenix. I would be immensely grateful if someone could kindly guide me on where to find this particular result within the Phoenix software.

Any assistance or insights you can provide would be greatly appreciated. Thank you very much for your time and assistance.

[cradhakr]

Hi Crystal,

Have you tried looking at “Core Output.” This can be found in Results Tab >> Text Output >> Core Output.

Thanks

Mouli

[bwendt]

Hi Crystal,

you can also inspect the BootOverall sheet which lists the detailed outcome from each bootstrap run:

The number of rows in that table divided by the number of replicates will give you the success rate.

Bernd

[crystal]

Thank you for your suggestion. I have tried looking at “Core Output,” but unfortunately, there were hundreds of thousands of lines to read, making it very difficult to find the information I need. I appreciate your help nonetheless.

@cradhakr

Hi Crystal,

Have you tried looking at “Core Output.” This can be found in Results Tab >> Text Output >> Core Output.

Thanks

Mouli

[crystal]

Thank you very much for your assistance; I have found the information you mentioned. However, I still have a question regarding the results. I noticed that the success rate appears to be consistently 100% when performing bootstraps in Phoenix. Is it genuinely possible to achieve a 100% success rate in bootstrap runs?

@bwendt_1

Hi Crystal,

you can also inspect the BootOverall sheet which lists the detailed outcome from each bootstrap run:

The number of rows in that table divided by the number of replicates will give you the success rate.

Bernd

[bwendt]

@crystal

Thank you very much for your assistance; I have found the information you mentioned. However, I still have a question regarding the results. I noticed that the success rate appears to be consistently 100% when performing bootstraps in Phoenix. Is it genuinely possible to achieve a 100% success rate in bootstrap runs?

The success rate depends on a couple of requirements:

1. a robust engine - with FOCE_ELS and QRPEM, Phoenix got two very strong and robust engines

2. a model that is representative of the whole training set, so that every re-combination of a training set that form a new bootstrap sample will converge

You might have been lucky to see a 100% success rate consistently, but eventually you will encounter a situation where a particular combination of data set and model will not achieve that maximum rate.

Bernd

[crystal]

Thank you so much, I really appreciate it.

The success rate depends on a couple of requirements:

1. a robust engine - with FOCE_ELS and QRPEM, Phoenix got two very strong and robust engines

2. a model that is representative of the whole training set, so that every re-combination of a training set that form a new bootstrap sample will converge

You might have been lucky to see a 100% success rate consistently, but eventually you will encounter a situation where a particular combination of data set and model will not achieve that maximum rate.

Bernd