Error in simulate_data() : object 'tau_location' not found

[mvuorre]

https://github.com/mvuorre/latent-mean-centering-ms/blob/224043dab2e9dc5c489d19680dd52d75d629c060/Generation_Code.r#L38

The data generation code (I turned it into a function) doesn't work currently as some variables are missing. @JoranTiU can you help with this?

My aim is to turn it into a function where we can vary the number of individuals and number of timepoints. We should be able to examine how the n_individuals and n_timepoints affect estimation in a typical observed mean centering vs latent mean centering models.

[JoranTiU]

I'm on it! 😊

[mvuorre]

Awesome, thank you. If it wouldn't be too much trouble, I think it would be nice if the simulated model was fairly simple and basically had two predictors: the lagged outcome and some other time varying covariate. Those are sufficient for highlighting the most important biases that can result from observed mean centering, and keeping the model simple would be good for our target audience (which we've yet to fully determine 😆). Having said that, you are the expert in the subject matter so feel free to do what you see best!

[JoranTiU]

No problem 😊. And I agree! That would be more than sufficient and will also make the message most clear (for whatever audience we choose 😝). I’ll start with a 2-variable VAR model, in which there are two lagged variables that also influence each other. I can then always update to have one-variable that is lagged, and another time-varying predictor that is linearly related to that one. That would give us a multilevel regression model with AR-residuals. That way we have two large classes of models that are often used with intensive longitudinal data 😊. Vriendelijke groet/Kind Regards, Joran Joran Jongerling | Assistant-Professor/Co-Director of TESC | Department of Methodology and Statistics | Tilburg School of Social and Behavioral Sciences | Reitse Poort, Room RP 15 | https://experiencesampling.nl/person/joran-jongerling/ ***@***.*** From: Matti Vuorre ***@***.***> Sent: donderdag 24 november 2022 9:27 To: mvuorre/latent-mean-centering-ms ***@***.***> Cc: Joran Jongerling ***@***.***>; Assign ***@***.***> Subject: Re: [mvuorre/latent-mean-centering-ms] Error in simulate_data() : object 'tau_location' not found (Issue #6) Awesome, thank you. If it wouldn't be too much trouble, I think it would be nice if the simulated model was fairly simple and basically had two predictors: the lagged outcome and some other time varying covariate. Those are sufficient for highlighting the most important biases that can result from observed mean centering, and keeping the model simple would be good for our target audience (which we've yet to fully determine 😆). Having said that, you are the expert in the subject matter so feel free to do what you see best! — Reply to this email directly, view it on GitHub<#6 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AVCJOPGRRV5ZM7AET7XO6ALWJ7FSNANCNFSM6AAAAAASJLGOZA>. You are receiving this because you were assigned.Message ID: ***@***.***>

[mvuorre]

That sounds excellent, thank you!

[JoranTiU]

👍😊 Vriendelijke groet/Kind Regards, Joran Joran Jongerling | Assistant-Professor/Co-Director of TESC | Department of Methodology and Statistics | Tilburg School of Social and Behavioral Sciences | Reitse Poort, Room RP 15 | https://experiencesampling.nl/person/joran-jongerling/ ***@***.*** From: Matti Vuorre ***@***.***> Sent: vrijdag 25 november 2022 10:28 To: mvuorre/latent-mean-centering-ms ***@***.***> Cc: Joran Jongerling ***@***.***>; Assign ***@***.***> Subject: Re: [mvuorre/latent-mean-centering-ms] Error in simulate_data() : object 'tau_location' not found (Issue #6) That sounds excellent, thank you! — Reply to this email directly, view it on GitHub<#6 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AVCJOPAFUZ7JNPVA54SXTHDWKCBCBANCNFSM6AAAAAASJLGOZA>. You are receiving this because you were assigned.Message ID: ***@***.******@***.***>>