invalid numlist has missing values Error in forestplot

[Seun Anjorin]

Hi,
Recently, when I use metan or admetan on aggregated data, there is always an error with the forestplot. It wills up the error below:

invalid numlist has missing values

Error in forestplot

Is anyone familiar with this error?

[Seun Anjorin]

I just tried with ipdover, I am having a

similar challenge generating forest plot. See cold and output below

[Seun Anjorin]

Please can anyone help me with this? This is the final stage of my analysis, I need to generate this forest plot

[Nick Cox]

I don't use any of these commands, so can't help much.

But #1 is puzzling. The screenshot shows first that Stata doesn't recognize forestplot and then that there is a problem in running it. In any case, you don't show the command you used.

[Seun Anjorin]

Thanks Nick and apology for not being explicit about the problem. Pls see below the command used and the outout


use http://fmwww.bc.edu/repec/bocode/m/metan_example_data, clear
(Example dataset for meta-analyses, Ross Harris 2006)


. metan tdeath tnodeath cdeath cnodeath, ///
> rd random label(namevar=id, yearvar=year) counts

Studies included: 20
Participants included: 3232

Meta-analysis pooling of Risk Differences
using the random-effects inverse-variance model
with DerSimonian-Laird estimate of tau²

---------------------------------------------------------------------
Study identifier |
(Year) | Risk Diff. [95% Conf. Interval] % Weight
----------------------+----------------------------------------------
AHS 60-80 (1984) | -0.122 -0.253 0.009 6.29
Ohno (1988) | 0.062 -0.073 0.197 6.10
CA,USA 81-85 (1992) | 0.026 -0.158 0.209 4.29
Walker (1992) | 0.036 -0.129 0.201 4.90
Ghadirian (1996) | 0.075 -0.114 0.265 4.10
Ewings (1996) | 0.050 -0.082 0.183 6.22
Hayes (1996) | 0.191 0.029 0.353 5.00
SWE 71-75 (1997) | -0.038 -0.415 0.338 1.39
Hartman (1998) | -0.110 -0.262 0.042 5.37
AHS (1998) | -0.138 -0.240 -0.036 7.75
Lee (1998) | 0.095 -0.205 0.396 2.05
Hsieh (1999) | 0.236 0.058 0.414 4.45
ATBC (2001) | 0.031 -0.171 0.233 3.77
CARET (2003) | 0.048 -0.067 0.164 7.04
Lifespan study (2004) | -0.072 -0.189 0.045 6.96
Joseph (2004) | 0.103 -0.106 0.311 3.61
Friedenreich (2004) | 0.041 -0.084 0.166 6.56
Allen (2004) | 0.182 -0.012 0.377 3.97
Weinstein (2005) | 0.036 -0.096 0.168 6.22
King (2005) | 0.040 -0.154 0.234 3.97
----------------------+----------------------------------------------
Overall, DL | 0.027 -0.020 0.074 100.00
---------------------------------------------------------------------

Test of overall effect = 0: z = 1.125 p = 0.261


Heterogeneity measures, calculated from the data
with Conf. Intervals based on Gamma (random-effects) distribution for Q
----------------------------------------------------------
Measure | Value df p-value
----------------------+-----------------------------------
Cochran's Q | 33.80 19 0.019
| -[95% Conf. Interval]-
H | 1.334 1.000 1.764
I² (%) | 43.8% 0.0% 67.9%
----------------------------------------------------------
H = relative excess in Cochran's Q over its degrees-of-freedom
I² = proportion of total variation in effect estimate due to between-study heterogeneity (based on Q)


Heterogeneity variance estimates
------------------------------------
Method | tau²
----------------------+-------------
DL | 0.0048
------------------------------------
invalid numlist has missing values

Error in forestplot
r(127);

end of do-file
r(127);


I used dataset and code from metan help file, I got the output above. This is the same challenge I have when I use my dataset.

Thanks

[Nick Cox]

That looks more answerable by an expert. David Fisher pops up here from time to time. I never met a better person to answer this kind of query.

[Seun Anjorin]

Thank you Nick. I am hoping to get some answers from David soon.

[David Fisher]

Hi Seun Anjorin

Could you confirm the results of typing the following into Stata?

. which metan
. which forestplot

The latest version is v4.03 28apr2021. Using this version under Stata MP/16.1, I am unable to replicate your error, I'm afraid.

I am aware that this error has occurred in the recent past, but I had hoped that it had been fixed! Hopefully, either updating or uninstalling/reinstalling will fix your issue -- but please do let me know if not!

Best wishes,

David.