Error when running x$mrtryx()

[DM087]

Hi @explodecomputer @yoonsucho

Thank you for the wonderful package. I was following the 'Guide to using MR-TRYX' on this page but I get an error

https://explodecomputer.github.io/tryx/articles/guide.html

`x$mrtryx() Using RadialMR package to detect outliers

Radial IVW

              Estimate  Std.Error   t value     Pr(>|t|)

Effect (Mod.2nd) 0.5991043 0.07718481 7.761945 8.363657e-15 Iterative 0.5991043 0.07718481 7.761945 8.363657e-15 Exact (FE) 0.6420699 0.04024276 15.954916 2.633294e-57 Exact (RE) 0.6101965 0.08849314 6.895410 5.042766e-11

Residual standard error: 1.92 on 232 degrees of freedom

F-statistic: 60.25 on 1 and 232 DF, p-value: 2.66e-13 Q-Statistic for heterogeneity: 855.6769 on 232 DF , p-value: 2.874833e-72

Outliers detected Number of iterations = 2 Error in if (radial$outliers[1] == "No significant outliers") { : the condition has length > 1`

Any thoughts on what could be going wrong here?

Thank you

[yoonsucho]

Hi,

Thanks for using our package. The error message you got indicate that there is no statistically meaningful outliers for your exposure and outcome. In this case mrtryx() would fail to proceed the further analyses after outlier detection. Perhaps you could try with other set of summary statistics for the exposure and/or the outcome. Hope this helps.

Thanks.

Best wishes, Yoonsu

On Thu, 31 Aug 2023 at 11:50, rs7412 @.***> wrote:

Hi @explodecomputer https://github.com/explodecomputer @yoonsucho https://github.com/yoonsucho

Thank you for the wonderful package. I was following the 'Guide to using MR-TRYX' on this page but I get an error

https://explodecomputer.github.io/tryx/articles/guide.html

`x$mrtryx() Using RadialMR package to detect outliers

Radial IVW

          Estimate  Std.Error   t value     Pr(>|t|)

Effect (Mod.2nd) 0.5991043 0.07718481 7.761945 8.363657e-15 Iterative 0.5991043 0.07718481 7.761945 8.363657e-15 Exact (FE) 0.6420699 0.04024276 15.954916 2.633294e-57 Exact (RE) 0.6101965 0.08849314 6.895410 5.042766e-11

Residual standard error: 1.92 on 232 degrees of freedom

F-statistic: 60.25 on 1 and 232 DF, p-value: 2.66e-13 Q-Statistic for heterogeneity: 855.6769 on 232 DF , p-value: 2.874833e-72

Outliers detected Number of iterations = 2 Error in if (radial$outliers[1] == "No significant outliers") { : the condition has length > 1`

Any thoughts on what could be going wrong here?

Thank you

— Reply to this email directly, view it on GitHub https://github.com/explodecomputer/tryx/issues/23, or unsubscribe https://github.com/notifications/unsubscribe-auth/AINL2A5ZOR4XAIH3RNKGWQ3XYBUALANCNFSM6AAAAAA4F37QIM . You are receiving this because you were mentioned.Message ID: @.***>

[DM087]

Hi Yoonsu,

Thank you.

I am trying to replicate the guide that you have provided here https://explodecomputer.github.io/tryx/articles/guide.html but it does not work.

In the guide it says 'Identified 11 outliers' but when I run it gives me the error

Error in if (radial$outliers[1] == "No significant outliers") { :
the condition has length > 1`

Thanks

[bluscrubs]

Also experiencing the same problem. In 'Guide to using MR-TRYX', output of x$mrtryx function reads

#> ...
#> F-statistic: 60.25 on 1 and 232 DF, p-value: 2.66e-13
#> Q-Statistic for heterogeneity: 855.6769 on 232 DF , p-value: 2.874833e-72
#> 
#>  Outliers detected 
#> Number of iterations = 2
#> **Warning in if (radial$outliers[1] == "No significant outliers") {: the condition
#> has length > 1** and only the first element will be used
#> Identified 11 outliers

However, when following the guide, I receive

library(tryx)
x <- Tryx$new(dat)
x$mrtryx()
# ...
# F-statistic: 60.25 on 1 and 232 DF, p-value: 2.66e-13
# Q-Statistic for heterogeneity: 855.6769 on 232 DF , p-value: 2.874833e-72
# 
# Outliers detected 
# Number of iterations = 2
# Error in if (radial$outliers[1] == "No significant outliers") { : 
#     the condition has length > 1

It seems the error is coming from the x$get_outliers function. Any idea how to troubleshoot?

Thank you so much for this package and your help!