search

yhoriuchi / projoint

A package for a more general, more straightforward, and more creative conjoint analysis
https://yhoriuchi.github.io/projoint/
Other
10 stars 0 forks source link

estimation without reshaping #20

Closed ataustin06 closed 1 year ago

ataustin06 commented 1 year ago

Hello! I have already reshaped my data. How can I run an error-corrected mm or amce estimation without using your reshaping function?

yhoriuchi commented 1 year ago

Hi @ataustin06 . Thank you for posting the question. I understand that some people have already reshaped the data using other packages or writing their own scripts. What is the unit of observation in your reshaped data? As long as it is respondent-task-id, then it is not very difficult to make a new function (by me) to generate an object with the projoint_data class.

ataustin06 commented 1 year ago

Yes, it is in the format of respondent - task - id. Specifically, I have id numbers for each respondent, a choice preference between two profiles, and id numbers for each experimental iteration. My data also includes traits for the profile chosen and not chosen, an identifier for the repeated experimental iteration, and respondent characteristics for estimating conditional amce's. Thank you for your response and your assistance!

yhoriuchi commented 1 year ago

Can you send me your data by email? (You don't need to post your data here). If you prefer, you can send me just the first 20-50 rows. I need to have a look at how your data frame is structured.

ataustin06 commented 1 year ago

Yes, I can send you the first few rows of the data scrubbed of identifiers.

aaronrkaufman commented 1 year ago

@yhoriuchi I think we'll need to add something to projoint so that it can also accept a data frame or tibble, and if so, turns that into a projoint_data object. This will require more arguments to projoint probably.

yhoriuchi commented 1 year ago

Yes. That's exactly what I am working on. Yusaku

On Mon, Aug 14, 2023 at 5:27 PM Aaron R. Kaufman @.***> wrote:

@yhoriuchi https://github.com/yhoriuchi I think we'll need to add something to projoint so that it can also accept a data frame or tibble, and if so, turns that into a projoint_data object. This will require more arguments to projoint probably.

— Reply to this email directly, view it on GitHub https://github.com/yhoriuchi/projoint/issues/20#issuecomment-1678091166, or unsubscribe https://github.com/notifications/unsubscribe-auth/AC332BFVGOAFLFKGRQBLVXLXVKJ33ANCNFSM6AAAAAA3O4A25M . You are receiving this because you were mentioned.Message ID: @.***>

yhoriuchi commented 1 year ago

I made a new function named make_projoint_data.R and added a new section (2.4 Read your already-wrangled tibble) to the second vignette. I now close this issue! Thank you, @ataustin06 , for your question. @aaronrkaufman , please check the Roxygen2 documentation and the vignettes.