[R-meta] Questions about model averaging with complex multilevel meta-analytic model
Viechtbauer, Wolfgang (NP)
wo||g@ng@v|echtb@uer @end|ng |rom m@@@tr|chtun|ver@|ty@n|
Mon Oct 3 11:54:00 CEST 2022
Happy to hear that. Please follow instructions carefully and don't just skip parts of the code.
I suspect dredge() is trying to fit a very large number of models. This could take a very long time. To figure out how many models there are, you can run:
res <- dredge(full_mod, evaluate=FALSE)
length(res)
If you then still want to go through with this, you might need to consider using parallelization (if you have access to a workstation with many cores or a cluster) as described here:
https://gist.github.com/wviechtb/891483eea79da21d057e60fd1e28856b
Best,
Wolfgang
>-----Original Message-----
>From: Margaret Slein [mailto:maslein using zoology.ubc.ca]
>Sent: Monday, 03 October, 2022 10:42
>To: Viechtbauer, Wolfgang (NP)
>Cc: r-sig-meta-analysis using r-project.org
>Subject: Re: [R-meta] Questions about model averaging with complex multilevel
>meta-analytic model
>
>1. Amazing, yes the "eval(metafor:::.MuMIn” did allow me to run your example and
>get model-averaged coefficient estimates! Wahoo!!
>
>However, when I tried running a dredge on my own model, after 10+ hours it had
>still not finished running.
>
>I’m wondering if this issue is because I am not model dredging, but using
>different model combinations by hand and trying to average those?
>
>2. Here is the session info:
>
>> sessionInfo()
>R version 4.2.1 (2022-06-23)
>Platform: x86_64-apple-darwin17.0 (64-bit)
>Running under: macOS Monterey 12.4
>
>Matrix products: default
>LAPACK:
>/Library/Frameworks/R.framework/Versions/4.2/Resources/lib/libRlapack.dylib
>
>locale:
>[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
>
>attached base packages:
>[1] stats graphics grDevices utils datasets methods base
>
>other attached packages:
> [1] AICcmodavg_2.3-
>1 devtools_2.4.4 usethis_2.1.6 corrplot_0.92 metaAidR_0.0.0.9
>000 glmulti_1.0.8
> [7] leaps_3.1 rJava_1.0-6 wesanderson_0.3.6 ape_5.6-
>2 viridis_0.6.2 viridisLite_0.4.1
>[13] MuMIn_1.47.1 rotl_3.0.12 patchwork_1.1.2 metafor_3.8-
>1 metadat_1.2-0 Matrix_1.5-1
>[19]
>forcats_0.5.2 stringr_1.4.1 dplyr_1.0.10 purrr_0.3.4 r
>eadr_2.1.2 tidyr_1.2.0
>[25] tibble_3.1.8 ggplot2_3.3.6 tidyverse_1.3.2
>
>loaded via a namespace (and not attached):
> [1] VGAM_1.1-7 googledrive_2.0.0 colorspace_2.0-
>3 ellipsis_0.3.2 fs_1.5.2 rstudioapi_0.14
> [7]
>remotes_2.4.2 bit64_4.0.5 fansi_1.0.3 lubridate_1.8.0 m
>athjaxr_1.6-0 xml2_1.3.3
>[13]
>splines_4.2.1 cachem_1.0.6 pkgload_1.3.0 jsonlite_1.8.0 b
>room_1.0.1 dbplyr_2.2.1
>[19]
>shiny_1.7.2 rentrez_1.2.3 compiler_4.2.1 httr_1.4.4 b
>ackports_1.4.1 assertthat_0.2.1
>[25]
>fastmap_1.1.0 gargle_1.2.1 cli_3.3.0 later_1.3.0 h
>tmltools_0.5.3 prettyunits_1.1.1
>[31]
>tools_4.2.1 gtable_0.3.1 glue_1.6.2 Rcpp_1.0.9 c
>ellranger_1.1.0 vctrs_0.4.1
>[37] nlme_3.1-
>157 ps_1.7.1 rvest_1.0.3 mime_0.12 miniUI_0.1
>.1.1 lifecycle_1.0.2
>[43] pacman_0.5.1 XML_3.99-0.10 googlesheets4_1.0.1 MASS_7.3-
>57 scales_1.2.1 vroom_1.5.7
>[49]
>hms_1.1.2 promises_1.2.0.1 parallel_4.2.1 curl_4.3.2 p
>bapply_1.5-0 memoise_2.0.1
>[55]
>gridExtra_2.3 stringi_1.7.8 pkgbuild_1.3.1 rlang_1.0.5 p
>kgconfig_2.0.3 rncl_0.8.6
>[61] lattice_0.20-
>45 htmlwidgets_1.5.4 bit_4.0.4 processx_3.7.0 tidyselect_1.1
>.2 plyr_1.8.7
>[67]
>magrittr_2.0.3 R6_2.5.1 generics_0.1.3 profvis_0.3.7 D
>BI_1.1.3 pillar_1.8.1
>[73] haven_2.5.1 withr_2.5.0 survival_3.3-
>1 modelr_0.1.9 crayon_1.5.1 utf8_1.2.2
>[79]
>tzdb_0.3.0 urlchecker_1.0.1 progress_1.2.2 grid_4.2.1 r
>eadxl_1.4.1 callr_3.7.2
>[85] reprex_2.0.2 digest_0.6.29 xtable_1.8-
>4 httpuv_1.6.6 unmarked_1.2.5 stats4_4.2.1
>[91] munsell_0.5.0 sessioninfo_1.2.2
>
>
><*)))><> <*)))><> <*)))><> <*)))><>
>
>Maggie Slein (she/her/hers)
>PhD Student, O’Connor Lab
>Department of Zoology
>Unceded xʷməθkʷəy̓əm (Musqueam) territory
>University of British Columbia
More information about the R-sig-meta-analysis
mailing list