Dear Statalist colleagues
Can someone help with what may be somewhat basic questions about metareg
or calculating information from the e store?
I am wanting output the results of metareg to an excel file.
This is to avoid error created by transferring information manually.
xml_tab does not do all that is wanted so I am looking to do something
else.
If I can calculate the wanted values from information stored in e then I
can accumulate
The wanted values from several runs of metareg and output the
information via xmlsave.
Information not returned immediately from the e store, and which have to
be derived are:
I2 (Isquared),
P-value(s) for trend(s)
P-value(s) for any constant(s).
P-value for Q
P-value for tau2
An example of a metareg command used is:
metareg Y X , wsse(Z) knapphartung reml lrtau2
Used with a very small dataset example (Y, X, Z) of interest, the
subsequent command ereturn list gives:
ereturn list
scalars:
e(N) = 5
e(tau2) = .1835974028291436
e(df_m) = 1
e(Q) = 9.833370680444618
e(df_Q) = 3
e(df_r) = 3
e(qKH) = .9240880161523819
e(remll) = -2.922534398865104
e(remll_c) = -4.774354927921407
e(chi2_c) = 3.703641058112606
macros:
e(cmd) : "metareg"
e(depvar) : "Y"
e(predict) : "metareg_p"
e(tau2_method) : "REML"
e(wsse) : "Z"
e(properties) : "b V"
matrices:
e(b) : 1 x 2
e(V) : 2 x 2
functions:
e(sample)
I can calculate certain P values as follows:
gen slope = _b[X]
gen slope_se =_se[X]
gen slope_t = slope/slope_se
gen slope_p = 2* ttail(df_r,abs(slope_t))
gen const =_b[_cons]
gen const_se =_se[_cons]
gen const_t = const/const_se
gen const_p = 2*ttail(df_r,abs(const_t))
gen Q_p = chi2tail(df_r,e(Q))
However, it is not clear to me how to calculate:
I2 - I can get this approximately via an inelegant route,
however I should like to know more and how to obtain this
information directly from that stored in e. How is this calculated? The
precise value given in the metareg immediately prior to the above
ereturn information is 0.695
P_Tau2 - Likewise, how is P for Tau2 derived from
information in the e store?
The precise value given in the metareg immediately prior to the above
ereturn information is 0.1836
Q_p - The value of P for Q I calculate is given by the formula above,
that is it works with this example but is the procedure correct? The
precise value given by metareg immediately prior to the above ereturn
information is 0.020.
I have a copy of the related technical bulletin on order. However it has
not yet arrived and may be beyond me to interpret it when it arrives.
I do hope someone is able to help.
With my thanks,
Geoff. Livesey
PS copy of metareg table:
Meta-regression Number of studies = 5
Fit of model without heterogeneity (tau2=0): Q (3 df) =
9.83337
Prob > Q = 0.020
Proportion of variation due to heterogeneity I-squared =
0.695
REML estimate of between-study variance: tau2 =
0.1836
Y Coef. Std. Err. t P>t [95%
Conf. Interval]
X .113846 .0372789 3.05 0.055 -.0047921
2324842
_cons 1.008651 .4453188 2.27 0.108
-.4085525 2.425854
Likelihood-ratio test of tau2=0: chibar2(01) = 3.70 Prob > chibar2 =
0.027
________________________________________________________________________
_____________________
Geoffrey Livesey B.Sc., Ph.D., R.P.H.Nutr.
Member of Virtual Consulting Group, Professional Biosciences
Consultants, Web site www.v-c-g.co.uk Member of SENSE. Professional
Nutrition Consultants. www.sense-nutrition.org.uk Registered Public
Health Nutritionist P294.
INDEPENDENT NUTRITION LOGIC (INLogic) Ltd
NUTRITION RESEARCH MANAGEMENT AND CONSULTANCY
VAT Reg. No. GB 731 9065 38
Registered in England No 4991400
Tel: +44-1953-606689
Mobile: +44-7990-964609
Fax: +44-1953-600218
E: glivesey@inlogic.co.uk
A: Pealerswell House
21 Bellrope Lane
Wymondham
Norfolk
NR18 OQX
United Kingdom
W: www.inlogic.co.uk
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/