Hello,
Thanks again for creating and maintaining this package. I was trying to recreate the plots in the JSS paper that accompanies this package using the exact code provided alongside the publication: https://www.jstatsoft.org/article/view/v043i13
However, when I run that code, the styling/colors of the plots come out slightly differently.
Paper plot:
Reproduced plot:
I see that the data are the same, but I just wondered what's going on with the coloring - was this an editorial change?
I also noticed that I cannot precisely replicate the results reported for summary(temp.unstab$ipw.weights) on pg. 14. Here is what I see:
> library(survival)
> library(splines)
> data(haartdat)
> unstabilized_weight_actual.survival <- ipwtm(exposure = haartind, family = "survival", denominator = ~ cd4.sqrt, id = patient, tstart = tstart, timevar = fuptime, type = "first", data = haartdat)
> unstabilized_weight_actual.survival.summary <- as.numeric(summary(unstabilized_weight_actual.survival$ipw.weights))
> unstabilized_weight_actual.survival.summary
[1] 1.002472 1.161205 1.372061 13.135058 15.415889 401.776343
The values are similar within a tolerance to those in the paper, but I would expect them to be identical. What do you think is going on here?
Thank you,
Arianna
> sessionInfo()
R version 4.6.0 (2026-04-24)
Platform: x86_64-pc-linux-gnu
Running under: Rocky Linux 8.10 (Green Obsidian)
Matrix products: default
BLAS: /opt/R/4.6.0/lib64/R/lib/libRblas.so
LAPACK: /opt/R/4.6.0/lib64/R/lib/libRlapack.so; LAPACK version 3.12.1
locale:
[1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8
[5] LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8 LC_PAPER=en_US.UTF-8 LC_NAME=C
[9] LC_ADDRESS=C LC_TELEPHONE=C LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C
time zone: Etc/UTC
tzcode source: system (glibc)
attached base packages:
[1] splines stats graphics grDevices utils datasets methods base
other attached packages:
[1] geepack_1.3.13 survival_3.8-6 ipw_1.3.0
loaded via a namespace (and not attached):
[1] Matrix_1.7-5 dplyr_1.2.1 compiler_4.6.0 brio_1.1.5 tidyselect_1.2.1
[6] tidyr_1.3.2 vdiffr_1.0.9 lattice_0.22-9 R6_2.6.1 generics_0.1.4
[11] MASS_7.3-65 backports_1.5.1 tibble_3.3.1 desc_1.4.3 nnet_7.3-20
[16] rprojroot_2.1.1 pillar_1.11.1 rlang_1.2.0 utf8_1.2.6 testthat_3.3.2
[21] broom_1.0.13 openval.snapshot_0.5.0 pkgload_1.5.2 otel_0.2.0 cli_3.6.6
[26] magrittr_2.0.5 grid_4.6.0 rstudioapi_0.18.0 lifecycle_1.0.5 nlme_3.1-169
[31] vctrs_0.7.3 glue_1.8.1 rsconnect_1.8.0 purrr_1.2.2 tools_4.6.0
[36] pkgconfig_2.0.3
Hello,
Thanks again for creating and maintaining this package. I was trying to recreate the plots in the JSS paper that accompanies this package using the exact code provided alongside the publication: https://www.jstatsoft.org/article/view/v043i13
However, when I run that code, the styling/colors of the plots come out slightly differently.
Paper plot:
Reproduced plot:
I see that the data are the same, but I just wondered what's going on with the coloring - was this an editorial change?
I also noticed that I cannot precisely replicate the results reported for
summary(temp.unstab$ipw.weights)on pg. 14. Here is what I see:The values are similar within a tolerance to those in the paper, but I would expect them to be identical. What do you think is going on here?
Thank you,
Arianna