explodecomputer / tryx

Other
10 stars 3 forks source link

Error when RadialMR does not detect any outliers #21

Open Schmytzi opened 3 years ago

Schmytzi commented 3 years ago

Summary

Calling Tryx$get_outliers() results in an error when RadialMR does not detect any outliers in the data that has been passed.

To Reproduce

library(tryx)
dat <- read.table("tryx_data.txt", sep="\t", stringsAsFactors=F, header=T)
x <- Tryx$new(dat)
x$get_outliers()

Expected result: Analysis should run without errors, reporting no outliers. Observed result:

Using RadialMR package to detect outliers

Radial IVW

                  Estimate  Std.Error  t value     Pr(>|t|)
Effect (Mod.2nd) 0.1824949 0.06214916 2.936403 0.0033204302
Iterative        0.1824975 0.06214849 2.936475 0.0033196508
Exact (FE)       0.1855602 0.04933755 3.761034 0.0001692127
Exact (RE)       0.1842288 0.07749245 2.377377 0.0549645010

Residual standard error: 1.26 on 6 degrees of freedom

F-statistic: 8.62 on 1 and 6 DF, p-value: 0.0261
Q-Statistic for heterogeneity: 9.527855 on 6 DF , p-value: 0.1459953

 Outliers detected
Number of iterations = 3
Error in `.rowNamesDF<-`(x, value = value) : invalid 'row.names' length

Attachments

tryx_data.txt