I am about to conduct a meta-regression analysis to fit some age-related changes. Since the GAMLSS approach is highly suggested in the normal regression of such data, I want to apply GAMLSS to my study. However, I could not find avalible packages to achieve this. Could any one help me? If indeed there's no available packages, is there any way to achieve that easily? I cannot simply apply the meta-regression with the normal GAMLSS method, since the latter does not take the variance of studies into account. Any suggestions are appreciated.
2
-
$\begingroup$ I cannot directly answer your question since I am not familiar enough with the gamlss package, but you can model non-linear relationships, for example using splines, using the metafor package; see here for some examples: metafor-project.org/doku.php/tips:non_linear_meta_regression $\endgroup$– WolfgangAug 23, 2022 at 15:56
-
1$\begingroup$ Thanks a lot for this suggestion! The spline regression is indeed a good alternative, if GAMLSS cannot be achieved. I'll see if it works good enough in my data. $\endgroup$– Jack YangAug 24, 2022 at 3:24
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
There is a meta-analysis example using gamlss on pages 341-345 (in the Chapter 10: Random effects) of
'Flexible Regression and Smoothing: Using GAMLSS in R’ M. D. Stasinopoulos, R. A. Rigby, G. Z. Heller, V. Voudouris and F. De Bastiani. Chapman and Hall/CRC, Boca Raton, 2017
