[R] help: forest plots

[Viechtbauer Wolfgang (STAT)]

If you have the bounds of a 95% CI in the form of:

exponential(estimate +/- 1.96*SE)

then it is easy to get the SE back:

SE = (log(UL) - log(LL)) / (2*1.96)

Then supply the estimates and the SEs to the forest() function:

forest(estimate, sei=SE)

You can use:

forest(estimate, sei=SE, transf=exp)

if you want to exponentiate the estimates and CI bounds. Alternatively, you can use:

forest(estimate, sei=SE, atransf=exp)

to exponentiate the x-axis values.

Best,

--
Wolfgang Viechtbauer                        http://www.wvbauer.com/
Department of Methodology and Statistics    Tel: +31 (0)43 388-2277
School for Public Health and Primary Care   Office Location:
Maastricht University, P.O. Box 616         Room B2.01 (second floor)
6200 MD Maastricht, The Netherlands         Debyeplein 1 (Randwyck)


----Original Message----
From: r-help-bounces at r-project.org
[mailto:r-help-bounces at r-project.org] On Behalf Of Kim Jung Hwa Sent:
Monday, December 14, 2009 05:22 To: C.H.
Cc: r-help at r-project.org
Subject: Re: [R] help: forest plots

> Thanks for this link. It helps, but I have to make lots of forest
> plots and these R scripts are not generic.
>
> Are you aware of similar functions as forest(), which can take input
> in the form of estimates, lower_limit, upper_limit? Thanks a lot!
>
> ~Kim
>
> On Sun, Dec 13, 2009 at 8:12 PM, C.H. <chainsawtiney at gmail.com> wrote:
>
>> This one does required the metafor package.
>>
>> http://tables2graphs.com/doku.php?id=04_regression_coefficients
>>
>>
>>
>> On Mon, Dec 14, 2009 at 8:13 AM, Kim Jung Hwa
>> <kimhwamaillist at gmail.com>
>> wrote:
>>> Hi All,
>>>
>>> I'm fitting a Poisson regression. And I want to plot 95% Confidence
>>> Interval of Regression Estimates.
>>>
>>> After coming back to original scale (using following formula):
>>>
>>> exponential(estimate +/- 1.96*SE),
>>>
>>> at best I can get the output in the form of estimates, lower_limit,
>>> upper_limit values.
>>>
>>> As far I know forest() in metafor package needs input in the form of
>>> estimates and their variances. In the above case can I still use
>>> forest()? OR if there exists some other function which takes such
>>> input and gives forest plots?
>>>
>>> Any help would be highly appreciated,
>>> Thanks,
>>> Kim