Hey, I'm doing analyses with binary logistic GEE. I was wondering if using
"independent" as working correlation structure takes into account the repeated measurements in the data or if this would be the same as using normal logistic regression and ignoring the fact that there are repeated measurements. (My goal is to correct for repeated measurements but I have no assumption regarding the correlation). Thanks!  Sent from: http://spssxdiscussion.1045642.n5.nabble.com/ ===================== To manage your subscription to SPSSXL, send a message to [hidden email] (not to SPSSXL), with no body text except the command. To leave the list, send the command SIGNOFF SPSSXL For a list of commands to manage subscriptions, send the command INFO REFCARD 
From the dialog box help... Working Correlation Matrix. This correlation matrix represents the withinsubject dependencies. Its size is determined by the number of measurements and thus the combination of values of withinsubject variables. You can specify one of the following structures:
On Mon, Dec 7, 2020 at 6:16 AM spssdummy <[hidden email]> wrote: Hey, I'm doing analyses with binary logistic GEE. I was wondering if using 
Thanks Jon:) However, I read through this before already and I'm still not
sure about my question and what choosing for 'independent' actually implies!  Sent from: http://spssxdiscussion.1045642.n5.nabble.com/ ===================== To manage your subscription to SPSSXL, send a message to [hidden email] (not to SPSSXL), with no body text except the command. To leave the list, send the command SIGNOFF SPSSXL For a list of commands to manage subscriptions, send the command INFO REFCARD 
All I can suggest is that you work through the GENLIN topic in the Algorithms doc. On Mon, Dec 7, 2020 at 8:22 AM spssdummy <[hidden email]> wrote: Thanks Jon:) However, I read through this before already and I'm still not 
Administrator

In reply to this post by spssdummy
Alternatively, you can try it and see what happens. I would work out an
example in SPSS if I had more time. But in all honesty, I find the code for estimating models with GEE much simpler in Stata, and therefore, was able to whip up this example quite quickly. To make things line up properly, you'll have to change the font to a fixed font such as Courier. (You may have to copy & paste into a text editor or word processor.) Compare the coefficients, SEs, z and pvalues and 95% CIs for Models 2 and 3. They are the same (to 3 decimals, at least). Model 2 was estimated using the xtgee command but with the corr(independent) option. Model 3 was estimated with the logit command, one of the commands that can be used to estimate ordinary binary logit models. HTH. . * Use a variation on the example shown here: . * https://www.stata.com/features/generalizedestimatingequations/ . clear . webuse nlswork (National Longitudinal Survey. Young Women 1426 years of age in 1968) . * Use logit link function rather than probit . * Model 1: GEE with corr(exchangeable) . xtgee union age not_smsa, i(idcode) /// > family(binomial) link(logit) corr(exchangeable) Iteration 1: tolerance = .08812485 Iteration 2: tolerance = .00597886 Iteration 3: tolerance = .00022492 Iteration 4: tolerance = 7.966e06 Iteration 5: tolerance = 2.747e07 GEE populationaveraged model Number of obs = 19,226 Group variable: idcode Number of groups = 4,150 Link: logit Obs per group: Family: binomial min = 1 Correlation: exchangeable avg = 4.6 max = 12 Wald chi2(2) = 29.83 Scale parameter: 1 Prob > chi2 = 0.0000  union  Coef. Std. Err. z P>z [95% Conf. Interval] + age  0.008 0.002 3.259 0.0011 0.003 0.013 not_smsa  0.250 0.056 4.482 0.0000 0.360 0.141 _cons  1.446 0.083 17.404 0.0000 1.609 1.284  . * Model 2: GEE with corr(independent) . xtgee union age not_smsa, i(idcode) /// > family(binomial) link(logit) corr(independent) Iteration 1: tolerance = 4.743e09 GEE populationaveraged model Number of obs = 19,226 Group variable: idcode Number of groups = 4,150 Link: logit Obs per group: Family: binomial min = 1 Correlation: independent avg = 4.6 max = 12 Wald chi2(2) = 102.32 Scale parameter: 1 Prob > chi2 = 0.0000 Pearson chi2(19226): 19227.17 Deviance = 20828.88 Dispersion (Pearson): 1.000061 Dispersion = 1.08337  union  Coef. Std. Err. z P>z [95% Conf. Interval] + age  0.010 0.003 3.726 0.0002 0.005 0.016 not_smsa  0.380 0.040 9.522 0.0000 0.458 0.301 _cons  1.409 0.089 15.804 0.0000 1.583 1.234  . * Model 3: An ordinary logit model . logit union age not_smsa Iteration 0: log likelihood = 10467.433 Iteration 1: log likelihood = 10414.653 Iteration 2: log likelihood = 10414.44 Iteration 3: log likelihood = 10414.44 Logistic regression Number of obs = 19,226 LR chi2(2) = 105.99 Prob > chi2 = 0.0000 Log likelihood = 10414.44 Pseudo R2 = 0.0051  union  Coef. Std. Err. z P>z [95% Conf. Interval] + age  0.010 0.003 3.726 0.0002 0.005 0.016 not_smsa  0.380 0.040 9.522 0.0000 0.458 0.301 _cons  1.409 0.089 15.804 0.0000 1.583 1.234  . end of dofile Here is the code, in case anyone wants it. * Use a variation on the example shown here: * https://www.stata.com/features/generalizedestimatingequations/ clear webuse nlswork * Use logit link function rather than probit * Model 1: GEE with corr(exchangeable) xtgee union age not_smsa, i(idcode) /// family(binomial) link(logit) corr(exchangeable) * Model 2: GEE with corr(independent) xtgee union age not_smsa, i(idcode) /// family(binomial) link(logit) corr(independent) * Model 3: An ordinary logit model logit union age not_smsa spssdummy wrote > Thanks Jon:) However, I read through this before already and I'm still not > sure about my question and what choosing for 'independent' actually > implies! > > > >  > Sent from: http://spssxdiscussion.1045642.n5.nabble.com/ > > ===================== > To manage your subscription to SPSSXL, send a message to > LISTSERV@.UGA > (not to SPSSXL), with no body text except the > command. To leave the list, send the command > SIGNOFF SPSSXL > For a list of commands to manage subscriptions, send the command > INFO REFCARD   Bruce Weaver [hidden email] http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." NOTE: My Hotmail account is not monitored regularly. To send me an email, please use the address shown above.  Sent from: http://spssxdiscussion.1045642.n5.nabble.com/ ===================== To manage your subscription to SPSSXL, send a message to [hidden email] (not to SPSSXL), with no body text except the command. To leave the list, send the command SIGNOFF SPSSXL For a list of commands to manage subscriptions, send the command INFO REFCARD

Bruce Weaver bweaver@lakeheadu.ca http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." NOTE: My Hotmail account is not monitored regularly. To send me an email, please use the address shown above. 
Free forum by Nabble  Edit this page 