Adding Bland- Altman statistical function

[linc50]

Pull Request

I have added the Bland-Altman sfunction, which can be used to assess the agreement between two quantitative methods of measurement. Please kindly help to review the code.

Thanks,
Chanjuan

[linc50]

close issue #850

[Melkiades]

Thanks @linc50 for working on this :)

We need a NEWS entry, the documentation is not present (you need to roxygenize and add description and fix the reference), and we need to have this statistical function is some tabulation example, test and list of stats (see get_stats). I would also consider moving this to a file that has similar statistical functions as the name itself does necessarily cut it for a self-explanatory file. Remember to use the styler btw ;)

[shajoezhu]

Hi @Melkiades , we will update this PR with a g_bland_altman function

[Melkiades]

Hi @Melkiades , we will update this PR with a g_bland_altman function

Ok super! so far, I think we have used the main user-facing (tables/plot) functions to name and direct files' content. I would probably wait to have the plot function and name it as you suggested, g_bland_altman. We can always refactor stats and methods to have a different semantic grouping at a later time (also after Daniels' work maybe?)

[shajoezhu]

lgtm! hey @Melkiades , I was wondering if you would want to take another look, Thanks

[Melkiades]

@shajoezhu, we can proceed with this, but I would appreciate a little clarification. There is no primary a_bland_altman function or some rtables-related function. I understand this could be part of a plan to add statistical functions for other packages or for other uses. If this is the case, I suggest we group them in a file (e.g. general_stats or external_stats) where we specify that they are for external uses; I expect more functions will gather there.

On the technical side, it is missing the function references in pkgdown and needs the plot function test.

[shajoezhu]

@shajoezhu, we can proceed with this, but I would appreciate a little clarification. There is no primary a_bland_altman function or some rtables-related function. I understand this could be part of a plan to add statistical functions for other packages or for other uses. If this is the case, I suggest we group them in a file (e.g. general_stats or external_stats) where we specify that they are for external uses; I expect more functions will gather there.

On the technical side, it is missing the function references in pkgdown and needs the plot function test.

Thanks @Melkiades , I created an issue #1175 to follow up on the ggplot test, let's do this after we resolve the svg plot testing issue.

the s_ function was dedicated for supporting the plot, there is no need for table at this moment, we had looked up on the standard before, and there was no example there.

I thought g_ pattern should allow us to add this to pkgdown site, correct me if I was wrong. probably no need to duplicate materials on the s_bland_altman function. thanks

[Melkiades]

@shajoezhu, we can proceed with this, but I would appreciate a little clarification. There is no primary a_bland_altman function or some rtables-related function. I understand this could be part of a plan to add statistical functions for other packages or for other uses. If this is the case, I suggest we group them in a file (e.g. general_stats or external_stats) where we specify that they are for external uses; I expect more functions will gather there.
On the technical side, it is missing the function references in pkgdown and needs the plot function test.

Thanks @Melkiades , I created an issue #1175 to follow up on the ggplot test, let's do this after we resolve the svg plot testing issue.
Thanks 👍

the s_ function was dedicated for supporting the plot, there is no need for table at this moment, we had looked up on the standard before, and there was no example there.
I see, then LGTM! 👍

I thought g_ pattern should allow us to add this to pkgdown site, correct me if I was wrong. probably no need to duplicate materials on the s_bland_altman function. thanks
Good point. It is true ^^, hence the absence of error in ci/cd

[Melkiades]

LGTM! Thanks for working on this ;)