[R-sig-ME] Advice for the R-package of lme4

[Ben Bolker]

   [Please keep r-sig-mixed-models using r-project.org in the Cc: list!]

   The warning message is *probably* irrelevant; to make it go away you 
would need to install the Matrix package from source on your machine 
(the binary packages on CRAN are built with the latest R version of a 
particular release series).

   A preliminary investigation on my side had the same general 
conclusions (fitting with 1.1-23 and 1.1-26 gave identical answers). I'm 
afraid that if you have tried with both versions of the package and 
can't replicate your former results exactly then, logically, *the 
lme4/lmerTest package versions cannot be the source of the problem*; 
some other aspect of your setup must have changed. Numerical 
computations are, unfortunately, subject to a great deal of 
(small-scale) instability due to changes in underlying linear algebra 
packages, compilers, etc..

  I'm not sure what else we can do to help.

   sincerely
    Ben Bolker

On 8/25/21 11:17 PM, 市川 淳 wrote:
> Thank you for your kind advice.
> 
> I installed the package of remotes and compilers.
> I tried to analyze using the combination of  lme4 1.1-20, 1.1-21, 
> 1.1-23, or 1.1-25 and  lmerTest 3.0-1 or lmerTest 3.1-3;
> however, I can't get the same results as before.
> 
> Incidentally, the following message is always displayed.
> 
> ====
> Warning message:
> Package 'Matrix' was built under R in version 3.6.3.
> //Actually, the message is in Japanese.
> =====
> 
> Very sorry to bother you while you are busy, but I appreciate your 
> cooperation.
> 
> Best regards
> Jun Ichikawa
> 
> 
> 2021年8月26日(木) 11:02 Ben Bolker <bbolker using gmail.com 
> <mailto:bbolker using gmail.com>>:
> 
>         If you want to check for yourself, you can retrieve older
>     versions of
>     the tarballs (.tar.gz files) from
>     https://cran.r-project.org/src/contrib/Archive/lme4/
>     <https://cran.r-project.org/src/contrib/Archive/lme4/>
> 
>        Alternatively, if you install the 'remotes' package, you can use
> 
>     remotes::install_version("lme4", "1.1-23")
> 
>     - this basically automates the process of downloading the tarball and
>     installing it.
> 
>        You *will* need development tools (compilers etc.) installed to do
>     this (an alternative is to use the 'checkpoint' package to retrieve old
>     package versions from MRAN (a versioned mirror of CRAN maintained by
>     Microsoft).
> 
> 
> 
>     On 8/25/21 9:10 PM, 市川 淳 wrote:
>      > Sorry for the sudden email.
>      > I would like you to give me your advice.
>      > This is my first post.
>      >
>      > I have often used the package of lme4 and lmerTest to conduct a
>     linear
>      > mixed model analysis.
>      > Meanwhile, I have been in trouble of whichthe values in the
>     output are
>      > slightly different between late April 2021 and August 2021.
>      > The source code, R environment, and csv files are the same
>     (R-3.6.1 in
>      > Windows 10).
>      >
>      > I surveyed and found out the following.
>      > Around June 2020, (maybe) I installed lme4 1.1-23 and lmerTest 3.0-1.
>      > I used these packages for a while; however I (maybe) updated lme4
>     1.1-26
>      > and lmerTest 3.1-3 a few days ago (in August 2021).
>      >
>      > I have checked the values of estimate, SE, p-value and so on in the
>      > output, and have noticed that these were different between the
>     packages
>      > mentioned above.
>      >
>      > Please advise me on the problem.
>      > I uploaded samples of the source code and csv file.
>      >
>     https://drive.google.com/drive/folders/1gbP-Ib-nW7-zrmUyrpuLmbODCuw3Zuot?usp=sharing
>     <https://drive.google.com/drive/folders/1gbP-Ib-nW7-zrmUyrpuLmbODCuw3Zuot?usp=sharing>
> 
>      >
>     <https://drive.google.com/drive/folders/1gbP-Ib-nW7-zrmUyrpuLmbODCuw3Zuot?usp=sharing
>     <https://drive.google.com/drive/folders/1gbP-Ib-nW7-zrmUyrpuLmbODCuw3Zuot?usp=sharing>>
>      >
>      > I look forward to hearing from you.
>      > Very sorry to bother you while you are busy, but I appreciate your
>      > cooperation.
>      >
>      > Best regards
>      > Anonymity X
> 
>     -- 
>     Dr. Benjamin Bolker
>     Professor, Mathematics & Statistics and Biology, McMaster University
>     Director, School of Computational Science and Engineering
>     Graduate chair, Mathematics & Statistics
> 

-- 
Dr. Benjamin Bolker
Professor, Mathematics & Statistics and Biology, McMaster University
Director, School of Computational Science and Engineering
Graduate chair, Mathematics & Statistics