Last updated on 2026-11-22 08:50:32 CET.
| Flavor | Version | Tinstall | Tcheck | Ttotal | Status | Flags |
|---|---|---|---|---|---|---|
| r-devel-linux-x86_64-debian-clang | 1.3-22 | 2.84 | 29.11 | 31.95 | OK | |
| r-devel-linux-x86_64-debian-gcc | 1.3-22 | 2.02 | 23.54 | 25.56 | ERROR | |
| r-devel-linux-x86_64-fedora-clang | 1.3-22 | 20.47 | OK | |||
| r-devel-linux-x86_64-fedora-gcc | 1.3-22 | 19.52 | OK | |||
| r-devel-windows-x86_64 | 1.3-22 | 3.00 | 49.00 | 52.00 | OK | |
| r-patched-linux-x86_64 | 1.3-22 | 2.69 | 24.73 | 27.42 | OK | |
| r-release-linux-x86_64 | 1.3-22 | 2.50 | 24.87 | 27.37 | OK | |
| r-release-macos-arm64 | 1.3-22 | 1.00 | 10.00 | 11.00 | OK | |
| r-release-macos-x86_64 | 1.3-22 | 2.00 | 30.00 | 32.00 | OK | |
| r-release-windows-x86_64 | 1.3-22 | 4.00 | 45.00 | 49.00 | OK | |
| r-oldrel-macos-arm64 | 1.3-22 | OK | ||||
| r-oldrel-macos-x86_64 | 1.3-22 | 2.00 | 48.00 | 50.00 | OK | |
| r-oldrel-windows-x86_64 | 1.3-22 | 5.00 | 48.00 | 53.00 | OK |
Version: 1.3-22
Check: tests
Result: ERROR
Running ‘BLUP.tests.R’ [1s/1s]
Running ‘OLS.r’ [0s/0s]
Running ‘predictionVariance.r’ [0s/0s]
Running ‘regress.tests.R’ [1s/1s]
Running ‘regressPaper.R’ [0s/1s]
Running the tests in ‘tests/predictionVariance.r’ failed.
Complete output:
> library(regress)
>
> ## Example of prediction variance computations
>
> ## Predictions
> x <- rnorm(15)
> x <- sort(x)
> y <- x + rnorm(15)
> predict(lm(y ~ x))
1 2 3 4 5 6 7 8
0.4880256 0.4898845 0.5002731 0.5034094 0.5043833 0.5160785 0.5269063 0.5521669
9 10 11 12 13 14 15
0.5562331 0.5709269 0.5747089 0.5773220 0.5965003 0.6045903 0.6169100
> new <- data.frame(x = x)
> predict(lm(y ~ x), new, se.fit = TRUE)
$fit
1 2 3 4 5 6 7 8
0.4880256 0.4898845 0.5002731 0.5034094 0.5043833 0.5160785 0.5269063 0.5521669
9 10 11 12 13 14 15
0.5562331 0.5709269 0.5747089 0.5773220 0.5965003 0.6045903 0.6169100
$se.fit
1 2 3 4 5 6 7 8
0.3191652 0.3125115 0.2770031 0.2669521 0.2639057 0.2306550 0.2070677 0.1926307
9 10 11 12 13 14 15
0.1964018 0.2222925 0.2315341 0.2384147 0.2982802 0.3270421 0.3733601
$df
[1] 13
$residual.scale
[1] 0.7362466
> pred.w.plim <- predict(lm(y ~ x), new, interval="prediction")
> pred.w.clim <- predict(lm(y ~ x), new, interval="confidence")
>
> model <- regress(y~x,verbose=10)
Extracting objects from call
Setting up kernel
Removing parts of random effects corresponding to missing values
Checking if we can apply the Sherman Morrison Woodbury identites for matrix inversion
1 sigma = 0.5052676 resid llik = -2.519528
1 adjusted sigma = 0.5420591
> ##plot(x,y)
> lm(y~x)$df.residual
[1] 13
> two <- qt((1-0.95)/2,13)
> hwid <- with(model,two*c(1,-1) %o% sqrt(predictedVariance2))
> pred.w.clim2 <- with(model,cbind(fitted,t(hwid)+as.vector(fitted)))
> hwid <- with(model,two*c(1,-1) %o% sqrt(predictedVariance2+predictedVariance))
> pred.w.plim2 <- with(model,cbind(fitted,t(hwid)+as.vector(fitted)))
>
> ## Test if predictions and prediction intervals are the same
> if(sum((pred.w.plim - pred.w.plim2)^2) + sum((pred.w.clim - pred.w.clim2^2)) > 1e-10) stop("Error with prediction variances")
Error: Error with prediction variances
Execution halted
Flavor: r-devel-linux-x86_64-debian-gcc