Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

SSE summarization #738

Open
3 tasks
barrettk opened this issue Feb 5, 2025 · 0 comments
Open
3 tasks

SSE summarization #738

barrettk opened this issue Feb 5, 2025 · 0 comments
Labels
needs SME input SME = Subject Matter Expert

Comments

@barrettk
Copy link
Collaborator

barrettk commented Feb 5, 2025

This issue is for discussing all design decisions related to the summarization step documented in #737

  • In the bootstrap, we compare the final parameter estimates of the based on model, to the bootstrap estimates:

    • Image
    • However in SSE, we want to compare the initial (or "true") parameter estimates of the based on model, to the sse estimates.
    • What do we call this column? We have “original” for the bootstrap. Tyler suggested "true value"
    • Do we export a new version of param_estimates_compare for doing this, or just do this internally for printing the sse summary
  • Does the information below capture everything we want, and in the format that we want? Note that run details, heuristics, and SSE specifications are also captured, just not printed below.

sse_sum$analysis_summary
> sse_sum$analysis_summary
# A tibble: 100 × 17
   absolute_model_path           run   error_msg termination_code THETA1 THETA2 THETA3  THETA4 THETA5 `SIGMA(1,1)` `OMEGA(1,1)` `OMEGA(2,1)` `OMEGA(2,2)`
   <chr>                         <chr> <lgl>                <dbl>  <dbl>  <dbl>  <dbl>   <dbl>  <dbl>        <dbl>        <dbl>        <dbl>        <dbl>
 1 /data/Projects/package_dev/b001   NA                       0   2.24  56.8   467.  -0.0787   4.14            1     0.0322              0       0.106 
 2 /data/Projects/package_dev/b002   NA                       0   2.23  55.2   439.  -0.0818   4.13            1     0.150               0       0.103 
 3 /data/Projects/package_dev/b003   NA                       0   2.26  53.8   410.  -0.0805   4.11            1     0.0343              0       0.142 
 4 /data/Projects/package_dev/b004   NA                       0  10.2    6.08   15.0 -0.0105   6.21            1     3.02                0      14.8   
 5 /data/Projects/package_dev/b005   NA                       0   2.32  55.4   423.  -0.0805   4.01            1     0.0937              0       0.301 
 6 /data/Projects/package_dev/b006   NA                       0   2.45  60.8   432.  -0.0910   4.65            1     0.000005            0       0.114 
 7 /data/Projects/package_dev/b007   NA                       0   2.15  56.5   446.  -0.0877   4.35            1     0.0633              0       0.102 
 8 /data/Projects/package_dev/b008   NA                     134   4.54  44.2    14.2 -0.0161   5.88            1     0.000005            0      13.3   
 9 /data/Projects/package_dev/b009   NA                       0   2.34  55.2   396.  -0.0819   4.03            1     0.116               0       0.0668
10 /data/Projects/package_dev/b010   NA                       0   2.33  56.4   457.  -0.0762   4.03            1     0.0488              0       0.234 
# ℹ 90 more rows
# ℹ 4 more variables: bbi_summary <list>, ofv <dbl>, condition_number <dbl>, any_heuristics <lgl>
# ℹ Use `print(n = ...)` to see more rows
sse_sum$parameter_estimates
> sse_sum$parameter_estimates
# A tibble: 900 × 10
   absolute_model_path                                    run   parameter_names estimate stderr random_effect_sd random_effect_sdse fixed diag  shrinkage
   <chr>                                                  <chr> <chr>              <dbl>  <dbl>            <dbl>              <dbl> <lgl> <lgl>     <dbl>
 1 /data/Projects/package_dev/bbr/inst/model/nonmem/basi001   THETA1            2.24       NA           NA                     NA FALSE NA           NA
 2 /data/Projects/package_dev/bbr/inst/model/nonmem/basi001   THETA2           56.8        NA           NA                     NA FALSE NA           NA
 3 /data/Projects/package_dev/bbr/inst/model/nonmem/basi001   THETA3          467.         NA           NA                     NA FALSE NA           NA
 4 /data/Projects/package_dev/bbr/inst/model/nonmem/basi001   THETA4           -0.0787     NA           NA                     NA FALSE NA           NA
 5 /data/Projects/package_dev/bbr/inst/model/nonmem/basi001   THETA5            4.14       NA           NA                     NA FALSE NA           NA
 6 /data/Projects/package_dev/bbr/inst/model/nonmem/basi001   OMEGA(1,1)        0.0322     NA            0.179                 NA FALSE TRUE         NA
 7 /data/Projects/package_dev/bbr/inst/model/nonmem/basi001   OMEGA(2,1)        0          NA            0                     NA TRUE  FALSE        NA
 8 /data/Projects/package_dev/bbr/inst/model/nonmem/basi001   OMEGA(2,2)        0.106      NA            0.325                 NA FALSE TRUE         NA
 9 /data/Projects/package_dev/bbr/inst/model/nonmem/basi001   SIGMA(1,1)        1          NA            1                     NA TRUE  TRUE         NA
10 /data/Projects/package_dev/bbr/inst/model/nonmem/basi002   THETA1            2.23       NA           NA                     NA FALSE NA           NA
# ℹ 890 more rows
# ℹ Use `print(n = ...)` to see more rows
@barrettk barrettk added the needs SME input SME = Subject Matter Expert label Feb 5, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
needs SME input SME = Subject Matter Expert
Projects
None yet
Development

No branches or pull requests

1 participant