Fit mjoint using multiple cores in parallel

[pradeepvirdee]

Hi Graeme,

I'm using the joineRML package to fit a joint model on a large dataset obtained from electronic patient records. I am building models on my eligible patient sample of ~80,000 males (~500,000 observations) and ~130,000 females (~650,000 observations) separately, I've been testing on smaller samples and found that larger sample sizes take exponentially longer. I intend to a use powerful computer/cluster. Is it possible for mjoint to use multiple cores in parallel to speed up the process?

I notice bootSE has built-in controls for this. Is this something mjoint has? If not, do you know of any workarounds to use multiple cores?

Thanks, Pradeep

[graemeleehickey]

mjoint() cannot be parellised; however, it can be made faster by using the quasi MC option. See here: https://core.ac.uk/download/pdf/323306058.pdf.

On Fri, 11 Dec 2020 at 11:53, pradeepvirdee notifications@github.com wrote:

Hi Graeme,

I'm using the joineRML package to fit a joint model on a large dataset obtained from electronic patient records. I am building models on my eligible patient sample of ~80,000 males (~500,000 observations) and ~130,000 females (~650,000 observations) separately, I've been testing on smaller samples and found that larger sample sizes take exponentially longer. I intend to a use powerful computer/cluster. Is it possible for mjoint to use multiple cores in parallel to speed up the process?

I notice bootSE has built-in controls for this. Is this something mjoint has? If not, do you know of any workarounds to use multiple cores?

Thanks, Pradeep

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/graemeleehickey/joineRML/issues/74, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABIG3QIWL4G4WP5AXME5IMLSUIBZ3ANCNFSM4UWQ2TPA .

[pradeepvirdee]

Thanks Graeme, I'll look into this.