The functions repeatedly fit the specified model, leaving out one observation/study at a time.

leave1out(x, …)

# S3 method for rma.uni
leave1out(x, digits, transf, targs, progbar=FALSE, …)
# S3 method for rma.mh
leave1out(x, digits, transf, targs, progbar=FALSE, …)
# S3 method for rma.peto
leave1out(x, digits, transf, targs, progbar=FALSE, …)

Arguments

x

an object of class "rma.mh", "rma.peto", or "rma.uni".

digits

integer specifying the number of decimal places to which the printed results should be rounded (if unspecified, the default is to take the value from the object).

transf

an optional argument specifying the name of a function that should be used to transform the model coefficients and interval bounds (e.g., transf=exp; see also transf). If unspecified, no transformation is used.

targs

optional arguments needed by the function specified under transf.

progbar

logical indicating whether a progress bar should be shown (the default is FALSE).

other arguments.

Details

The model specified via x must be a model without moderators (i.e., either a fixed- or a random-effects model and not a fixed-effects with moderators or mixed-effects model).

Value

An object of class "list.rma". The object is a list containing the following components:

estimate

estimated coefficients of the model.

se

standard errors of the coefficients.

zval

test statistics of the coefficients.

pval

p-values for the test statistics.

ci.lb

lower bounds of the confidence intervals for the coefficients.

ci.ub

upper bounds of the confidence intervals for the coefficients.

Q

test statistics for the tests of heterogeneity.

Qp

p-values for the tests of heterogeneity.

tau2

estimated amounts of (residual) heterogeneity (only for random-effects models).

I2

values of \(I^2\) (only for random-effects models).

H2

values of \(H^2\) (only for random-effects models).

The "list.rma" object is formatted and printed with print.list.rma.

Note

When using the transf option, the transformation is applied to the estimated coefficients and the corresponding interval bounds. The standard errors are then set equal to NA and are omitted from the printed output.

References

Viechtbauer, W. (2010). Conducting meta-analyses in R with the metafor package. Journal of Statistical Software, 36(3), 1--48. http://www.jstatsoft.org/v36/i03/.

Viechtbauer, W., & Cheung, M. W.-L. (2010). Outlier and influence diagnostics for meta-analysis. Research Synthesis Methods, 1, 112--125.

See also

Examples

### calculate log risk ratios and corresponding sampling variances dat <- escalc(measure="RR", ai=tpos, bi=tneg, ci=cpos, di=cneg, data=dat.bcg) ### random-effects model res <- rma(yi, vi, data=dat) ### leave-one-out analysis leave1out(res)
#> #> estimate se zval pval ci.lb ci.ub Q Qp tau2 #> 1 -0.7071 0.1900 -3.7223 0.0002 -1.0794 -0.3348 151.5826 0.0000 0.3362 #> 2 -0.6540 0.1807 -3.6195 0.0003 -1.0082 -0.2999 145.3176 0.0000 0.2926 #> 3 -0.6856 0.1857 -3.6916 0.0002 -1.0495 -0.3216 150.1970 0.0000 0.3207 #> 4 -0.6284 0.1766 -3.5580 0.0004 -0.9746 -0.2822 96.5626 0.0000 0.2628 #> 5 -0.7642 0.1918 -3.9845 0.0001 -1.1401 -0.3883 151.3200 0.0000 0.3278 #> 6 -0.7109 0.2003 -3.5499 0.0004 -1.1034 -0.3184 128.1867 0.0000 0.3596 #> 7 -0.6552 0.1805 -3.6307 0.0003 -1.0090 -0.3015 145.8296 0.0000 0.2930 #> 8 -0.7948 0.1799 -4.4184 0.0000 -1.1473 -0.4422 67.9858 0.0000 0.2732 #> 9 -0.7412 0.1967 -3.7686 0.0002 -1.1267 -0.3557 152.2051 0.0000 0.3495 #> 10 -0.6530 0.1843 -3.5439 0.0004 -1.0142 -0.2919 139.8271 0.0000 0.2987 #> 11 -0.7579 0.1958 -3.8708 0.0001 -1.1416 -0.3741 151.4655 0.0000 0.3405 #> 12 -0.7598 0.1821 -4.1727 0.0000 -1.1167 -0.4029 150.7868 0.0000 0.3082 #> 13 -0.7775 0.1855 -4.1908 0.0000 -1.1412 -0.4139 149.7884 0.0000 0.3037 #> I2 H2 #> 1 93.2259 14.7622 #> 2 92.2540 12.9098 #> 3 92.9354 14.1551 #> 4 90.4125 10.4302 #> 5 92.7634 13.8187 #> 6 90.9118 11.0033 #> 7 92.2777 12.9495 #> 8 87.0314 7.7109 #> 9 93.2133 14.7346 #> 10 92.2322 12.8737 #> 11 91.8110 12.2114 #> 12 92.6782 13.6579 #> 13 92.3444 13.0623 #>
leave1out(res, transf=exp)
#> #> estimate zval pval ci.lb ci.ub Q Qp tau2 I2 H2 #> 1 0.4931 -3.7223 0.0002 0.3398 0.7155 151.5826 0.0000 0.3362 93.2259 14.7622 #> 2 0.5199 -3.6195 0.0003 0.3649 0.7409 145.3176 0.0000 0.2926 92.2540 12.9098 #> 3 0.5038 -3.6916 0.0002 0.3501 0.7250 150.1970 0.0000 0.3207 92.9354 14.1551 #> 4 0.5334 -3.5580 0.0004 0.3774 0.7541 96.5626 0.0000 0.2628 90.4125 10.4302 #> 5 0.4657 -3.9845 0.0001 0.3198 0.6782 151.3200 0.0000 0.3278 92.7634 13.8187 #> 6 0.4912 -3.5499 0.0004 0.3318 0.7273 128.1867 0.0000 0.3596 90.9118 11.0033 #> 7 0.5193 -3.6307 0.0003 0.3646 0.7397 145.8296 0.0000 0.2930 92.2777 12.9495 #> 8 0.4517 -4.4184 0.0000 0.3175 0.6426 67.9858 0.0000 0.2732 87.0314 7.7109 #> 9 0.4765 -3.7686 0.0002 0.3241 0.7007 152.2051 0.0000 0.3495 93.2133 14.7346 #> 10 0.5205 -3.5439 0.0004 0.3627 0.7469 139.8271 0.0000 0.2987 92.2322 12.8737 #> 11 0.4687 -3.8708 0.0001 0.3193 0.6879 151.4655 0.0000 0.3405 91.8110 12.2114 #> 12 0.4678 -4.1727 0.0000 0.3274 0.6684 150.7868 0.0000 0.3082 92.6782 13.6579 #> 13 0.4595 -4.1908 0.0000 0.3194 0.6611 149.7884 0.0000 0.3037 92.3444 13.0623 #>
### meta-analysis of the (log) risk ratios using the Mantel-Haenszel method res <- rma.mh(measure="RR", ai=tpos, bi=tneg, ci=cpos, di=cneg, data=dat.bcg) ### leave-one-out analysis leave1out(res)
#> #> estimate se zval pval ci.lb ci.ub Q Qp #> 1 -0.4514 0.0394 -11.4462 0.0000 -0.5287 -0.3741 151.9153 0.0000 #> 2 -0.4410 0.0395 -11.1512 0.0000 -0.5185 -0.3635 145.5727 0.0000 #> 3 -0.4495 0.0394 -11.4001 0.0000 -0.5267 -0.3722 150.5130 0.0000 #> 4 -0.3392 0.0414 -8.1874 0.0000 -0.4205 -0.2580 96.5629 0.0000 #> 5 -0.4614 0.0400 -11.5474 0.0000 -0.5397 -0.3831 151.6611 0.0000 #> 6 -0.3666 0.0448 -8.1809 0.0000 -0.4544 -0.2788 129.2200 0.0000 #> 7 -0.4466 0.0395 -11.3046 0.0000 -0.5241 -0.3692 146.2134 0.0000 #> 8 -0.7758 0.0520 -14.9267 0.0000 -0.8777 -0.6739 68.3763 0.0000 #> 9 -0.4532 0.0399 -11.3633 0.0000 -0.5314 -0.3751 152.5497 0.0000 #> 10 -0.4278 0.0399 -10.7338 0.0000 -0.5059 -0.3497 140.0446 0.0000 #> 11 -0.4698 0.0421 -11.1706 0.0000 -0.5522 -0.3874 151.8140 0.0000 #> 12 -0.4566 0.0394 -11.5865 0.0000 -0.5338 -0.3794 151.1253 0.0000 #> 13 -0.4637 0.0398 -11.6542 0.0000 -0.5417 -0.3858 150.1246 0.0000 #>
leave1out(res, transf=exp)
#> #> estimate zval pval ci.lb ci.ub Q Qp #> 1 0.6367 -11.4462 0.0000 0.5894 0.6879 151.9153 0.0000 #> 2 0.6434 -11.1512 0.0000 0.5954 0.6952 145.5727 0.0000 #> 3 0.6380 -11.4001 0.0000 0.5905 0.6892 150.5130 0.0000 #> 4 0.7123 -8.1874 0.0000 0.6568 0.7726 96.5629 0.0000 #> 5 0.6304 -11.5474 0.0000 0.5829 0.6818 151.6611 0.0000 #> 6 0.6931 -8.1809 0.0000 0.6348 0.7567 129.2200 0.0000 #> 7 0.6398 -11.3046 0.0000 0.5921 0.6913 146.2134 0.0000 #> 8 0.4603 -14.9267 0.0000 0.4158 0.5097 68.3763 0.0000 #> 9 0.6356 -11.3633 0.0000 0.5878 0.6872 152.5497 0.0000 #> 10 0.6520 -10.7338 0.0000 0.6030 0.7049 140.0446 0.0000 #> 11 0.6251 -11.1706 0.0000 0.5757 0.6788 151.8140 0.0000 #> 12 0.6334 -11.5865 0.0000 0.5864 0.6843 151.1253 0.0000 #> 13 0.6289 -11.6542 0.0000 0.5817 0.6799 150.1246 0.0000 #>
### meta-analysis of the (log) odds ratios using Peto's method res <- rma.peto(ai=tpos, bi=tneg, ci=cpos, di=cneg, data=dat.bcg) ### leave-one-out analysis leave1out(res)
#> #> estimate se zval pval ci.lb ci.ub Q Qp #> 1 -0.4722 0.0408 -11.5791 0.0000 -0.5521 -0.3923 167.2005 0.0000 #> 2 -0.4616 0.0409 -11.2751 0.0000 -0.5418 -0.3814 160.5154 0.0000 #> 3 -0.4702 0.0408 -11.5317 0.0000 -0.5501 -0.3903 165.7913 0.0000 #> 4 -0.3591 0.0435 -8.2533 0.0000 -0.4443 -0.2738 112.1274 0.0000 #> 5 -0.4832 0.0413 -11.6873 0.0000 -0.5642 -0.4021 166.3727 0.0000 #> 6 -0.3710 0.0451 -8.2328 0.0000 -0.4593 -0.2827 139.3918 0.0000 #> 7 -0.4660 0.0408 -11.4343 0.0000 -0.5459 -0.3861 158.6048 0.0000 #> 8 -0.8161 0.0531 -15.3842 0.0000 -0.9201 -0.7122 67.1837 0.0000 #> 9 -0.4747 0.0413 -11.4973 0.0000 -0.5557 -0.3938 167.7284 0.0000 #> 10 -0.4486 0.0413 -10.8488 0.0000 -0.5296 -0.3675 155.8992 0.0000 #> 11 -0.4911 0.0434 -11.3135 0.0000 -0.5762 -0.4060 166.5326 0.0000 #> 12 -0.4775 0.0407 -11.7236 0.0000 -0.5573 -0.3976 166.0703 0.0000 #> 13 -0.4853 0.0411 -11.7960 0.0000 -0.5659 -0.4046 164.7427 0.0000 #>
leave1out(res, transf=exp)
#> #> estimate zval pval ci.lb ci.ub Q Qp #> 1 0.6236 -11.5791 0.0000 0.5757 0.6755 167.2005 0.0000 #> 2 0.6303 -11.2751 0.0000 0.5817 0.6829 160.5154 0.0000 #> 3 0.6249 -11.5317 0.0000 0.5769 0.6769 165.7913 0.0000 #> 4 0.6983 -8.2533 0.0000 0.6413 0.7605 112.1274 0.0000 #> 5 0.6168 -11.6873 0.0000 0.5688 0.6689 166.3727 0.0000 #> 6 0.6900 -8.2328 0.0000 0.6317 0.7538 139.3918 0.0000 #> 7 0.6275 -11.4343 0.0000 0.5793 0.6797 158.6048 0.0000 #> 8 0.4421 -15.3842 0.0000 0.3985 0.4906 67.1837 0.0000 #> 9 0.6220 -11.4973 0.0000 0.5737 0.6745 167.7284 0.0000 #> 10 0.6385 -10.8488 0.0000 0.5888 0.6924 155.8992 0.0000 #> 11 0.6120 -11.3135 0.0000 0.5621 0.6663 166.5326 0.0000 #> 12 0.6204 -11.7236 0.0000 0.5728 0.6719 166.0703 0.0000 #> 13 0.6155 -11.7960 0.0000 0.5679 0.6672 164.7427 0.0000 #>