Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

From |
"Lukas Bösch" <L.Boesch@gmx.de> |

To |
statalist@hsphsun2.harvard.edu |

Subject |
Re: st: xtmixed with nonrtolerance. What happens? |

Date |
Fri, 24 Jun 2011 19:20:58 +0200 |

-------- Original-Nachricht -------- > Datum: Fri, 24 Jun 2011 10:24:19 -0400 > Von: Joerg Luedicke <joerg.luedicke@gmail.com> > An: statalist@hsphsun2.harvard.edu > Betreff: Re: st: xtmixed with nonrtolerance. What happens? > See my comments below: Thank you for your comments. They make me think exactly about what I am doing, although it doesnt seem as bad to me as you think it is... > > On Fri, Jun 24, 2011 at 5:53 AM, "Lukas Bösch" <L.Boesch@gmx.de> wrote: > > Thank you for your help. I am writing my diploma thesis about the > influence of socioeconomic factors on the export of wild species. It took me a > long time to recode and transform the original dataset, which comprised about > 1 million observations, to a final compact dataset and i did not expect to > spend a lot of time with the calculation of the models. But as always, > something doesn't work...The work is due end of july and i really didnt expect > to take two weeks just to calculate the modells. > > > > 1) Concerning the missings, i took care to delete all data i don't have > the complete time series for. This means that i had to drop from 130 > countries over a 20 years period to 40 countries over a 15 years period. On the > other hand, there are deffinitely no missing values. > > > > This is neither necessary nor is it a good idea. But given your time > constraints, it is probably more important to worry about the other, > perhaps more severe problems with your data. > I have been told by my advisor that i need an observation in the independent variables for all observations in the dependent variable. I would therefor need a complete dataset without any missings. My idea was to include as much as possible to have exports from all different kinds of countries, but i did as i was told to. > > 2) It is the first time i am dealing with mixed modells, so i am not > sure about the terminology and the interpretation of the randome effects. In > my opinion the structure is not hierarchic as no random effects is nested in > another one, like in the always cited example: Students (level one) are > nested in classes (level two) and classes nested in schools (level three). > This is how i understand the hierarchic structure. In my dataset however, > genus is not nested in countries, as the same genus can be exportet by many > countries, which is the case for a lot of genus like falco, ursus or > crocodylus. This is why i defined the random effects the following way: || _all: > R.country || _all: R.genus > > > > The terminology indeed varies across textbooks. However, you have > observations nested in countries and observations nested in -genus- > (though, I am not 100% convinced that the latter is actually true). > There is just no hierarchical relation between country and genus. > > > 3) The dependent variable contains decimals because i had to recode and > transform it. In the original dataset, the export was classified in > species, categories and units. One countrie can export, products of brown bears, > grizzly bears, panda bears and so on (13756 species)... Then the country can > export living bears, dead bears, bear fur, bear bones, bear gall, bear > teeth and so on (89 categories)... Finally the country can export kg, m, g, > units and so on of a given product (24 units). This means that canada exports > about 178 different bear products a year in the original dataset. I > weighted, deleted and transformed the data, according to theoretical, practical > and pragmactic thoughts in order to get one exported product for each genus. > In the final dataset, canada exports one bear product a year. > > > > I don't understand this at all: Canada exports 178 bear products, but > in your data Canada exports 1 bear product. This makes no sense to me. > Ok, maybe it is more comprehensible with a hypothetical example: In 1991 canada exported in the original data 1 bear skull, 0 bear bones, 100 kg bear meat, 2 bear furs and 0 bear gall. It has 5 bear export categories. Maybe this is the reason for the missunderstanding. When i wrote that Indonesia had 70 exports I didnt mean that indonesia exported 70 items but that each year, indonesia has 70 export categories, which means 1050 observations for india. In some years, those categories dont get exported at all, whereas in other years the do. In 1991 canada exported in my final data 3 bear products. There is only one bear export category > > 4) If every country had one export a year, then there would be 640 > observations. Some countries, however, have many exports a year. For example, > Indonesia has 70 exports a year. It exports 70 different genus a year. If i > would sum up those 70 export to one export, the whole modell would be much > easier, but i thought the mixed model approach was conceived for this kind > of problems. > You are saying: "Indonesia has 70 exports a year. It exports 70 > different genus a year." This confuses me again: I would think if they > export 70 genuses they have _at least_ 70 exports, in the probably > special case that they exported exactly 1 unit per genus. So are you > counting the number of exports or the number of different genuses of > which exemplars are getting exported? And: if you are only including > positive realizations of genus (i.e, genuses for which exports are > > 0), where do your zeros come from? > I am counting the number of different genuses of which exemplars are getting exported at least once in the time period. This is where all the zeroes come from. If for example indonesia only exports in 2005 6 pineapple corals, then in all the other years there is a zero in indonesias pineapple export > > > > 5) It is thrue that there are a lot of zeroes in my dependent variable. > In the original dataset, there were so many, that i knew this would become > a problem. After transforming the dataset to my final version there were > still a lot of zeroes, in this case, 1 stands for zeroes. > > > > null | Freq. Percent Cum. > > ------------+----------------------------------- > > 0 | 9,501 24.32 24.32 > > 1 | 29,571 75.68 100.00 > > ------------+----------------------------------- > > Total | 39,072 100.00 > > > > This is why i decided to analyze only the genus where an export took > part in half of the years: > > This is problematic again. If you throw out countries that are not > exporting any wild animals you are probably imposing some heavy duty > selection bias. My original plan was to recode the values > 0 to one and to make a logistic analysis to see if a specific species gets exported or not. At this step, stata said r(1400) numerical overflow Then in the second step to throw out all the 0 and see in a linear model what influences the amount that gets exported. As i encountered the r(1400) message i thought i should try another approach. > > > > > > Variable | Obs Mean Std. Dev. Min > Max > > -------------+-------------------------------------------------------- > > quantity | 6192 8545.829 116704.3 0 > 4019264 > > > > > So, which is the country that exports 4019264 wild animals in one year? > In 1994 the Netherland exported 4019264 snowdrops. These are flowers and also registered in the CITES Appendix > > > I am going to try the poisson regression if this is better to analyze > > data with lot of zeroes. I didn't expect however that the final data > still had to many zeroes. > > > > > Before you start with any kind of model, I would suggest getting your > data straight. What are observational level units? Is the dependent > variable a count variable? Do the values and distribution of the > dependent variable make sense? And so on. I also strongly suggest that > you discuss these matters with your advisor! > > J. This is what i will do next week. Althouhg I allready discussed all those steps with him. He approved for example my idea to analyze only those genus where an export took part in half of the years. I will print the emails i had with you and others from statalist and see what he thinks. Thank you Lukas > > > > > -------- Original-Nachricht -------- > >> Datum: Thu, 23 Jun 2011 22:56:54 -0400 > >> Von: Joerg Luedicke <joerg.luedicke@gmail.com> > >> An: statalist@hsphsun2.harvard.edu > >> Betreff: Re: st: xtmixed with nonrtolerance. What happens? > > > >> A couple of points: > >> > >> 1) You surely assume a hierarchical structure: you have observations > >> at Level-1, cross-classified across 2 Level-2 factors. > >> > >> 2) It is still not clear what your unit of analysis is? What are those > >> 6192? They cannot be yearly observations within country as that would > >> only result in 640 observations. > >> > >> 3) If your dependent variable is supposed to represent the number of > >> exports (in a given year?), why does it contain decimals and not only > >> integers? > >> > >> 4) Have you ever spent some time looking at the distribution of your > >> dependent variable? When you standardize it, it ranges from .07 > >> standard deviations below the mean to 34 standard deviations above the > >> mean!! My guess is that you are looking at some crazy distribution > >> like this: > >> > >> gen x=rgamma(.001,100000) > >> > >> with some very high values but with the majority of values being zeros > >> or close to zero. I suspect that there is either something wrong with > >> this variable or with your entire data set-up. > >> > >> 5) If it turns out that everything in your data is correct, then > >> trying to fit a linear model to these data is certainly the wrong > >> approach. > >> > >> > >> Joerg > >> > >> > >> On Thu, Jun 23, 2011 at 4:40 PM, "Lukas Bösch" <L.Boesch@gmx.de> > wrote: > >> > Because i didnt transform the year and the export, named as quantity, > >> into z-scores they kept their original names in the first models. > >> > I just did the transformation and ran the model again, but it still > >> doesnt converge, however seems to work a little better. > >> > > >> > . xtmixed centquantity2 centyear2 centforestarea2 centgdp2 > centlandarea2 > >> centpopulation2|| _all: R.country || _all: R.genus > >> > > >> > Performing EM optimization: > >> > Performing gradient-based optimization: > >> > > >> > Iteration 0: log restricted-likelihood = -4875.1075 > >> > Iteration 1: log restricted-likelihood = -4870.6476 > >> > Iteration 2: log restricted-likelihood = -4870.5095 > >> > Iteration 3: log restricted-likelihood = -4870.4438 > >> > Iteration 4: log restricted-likelihood = -4870.4118 (backed up) > >> > Iteration 5: log restricted-likelihood = -4870.4039 (backed up) > >> > Iteration 6: log restricted-likelihood = -4870.3999 (backed up) > >> > Iteration 7: log restricted-likelihood = -4870.3979 (backed up) > >> > Iteration 8: log restricted-likelihood = -4870.3969 (backed up) > >> > Iteration 9: log restricted-likelihood = -4870.3967 (backed up) > >> > numerical derivatives are approximate > >> > nearby values are missing > >> > Iteration 10: log restricted-likelihood = -4870.3966 (backed up) > >> > numerical derivatives are approximate > >> > nearby values are missing > >> > Iteration 11: log restricted-likelihood = -4870.3966 (backed up) > >> > numerical derivatives are approximate > >> > nearby values are missing > >> > numerical derivatives are approximate > >> > nearby values are missing > >> > Hessian has become unstable or asymmetric > >> > > >> > Mixed-effects REML regression Number of > obs > >> = 6192 > >> > Group variable: _all Number > of > >> groups = 1 > >> > > >> Obs per group: min = 6192 > >> > > >> avg = 6192.0 > >> > > >> max = 6192 > >> > > >> Wald chi2(5) = 9.26 > >> > Log restricted-likelihood = -4875.1075 Prob > chi2 > >> = 0.0991 > >> > > >> > centquanti~2 | Coef. Std. Err. z P>|z| > >> [95% Conf. Interval] > >> > centyear2 | -.0169763 .008528 -1.99 0.047 > >> -.033691 -.0002616 > >> > centfores~a2 | -.0846178 .0568262 -1.49 0.136 > >> -.1959951 .0267595 > >> > centgdp2 | -.0173484 .0354612 -0.49 0.625 > >> -.0868509 .0521542 > >> > centlandar~2 | -.4531947 .5468347 -0.83 0.407 > >> -1.524971 .6185816 > >> > centpopul~n2 | .1910553 .0876979 2.18 0.029 > >> .0191707 .36294 > >> > _cons | .2434439 .4596746 0.53 0.596 > >> -.6575018 1.14439 > >> > > >> > Random-effects Parameters | Estimate Std. Err. [95% > >> Conf. Interval] > >> > _all: Identity | > >> > sd(R.country) | 2.684813 . > >> > _all: Identity | > >> > sd(R.genus) | .0579011 . > >> > sd(Residual) | .5155702 . > >> > LR test vs. linear regression: chi2(2) = 7810.42 Prob > > >> chi2 = 0.0000 > >> > > >> > Note: LR test is conservative and provided only for reference. > >> > Warning: convergence not achieved; estimates are based on iterated EM > >> > > >> > > >> > > >> > Here the summarize output of all the variables: > >> > > >> > . sum centquantity2 > >> > > >> > Variable | Obs Mean Std. Dev. > Min > >> Max > >> > > -------------+-------------------------------------------------------- > >> > centquanti~2 | 6192 2.17e-09 1 > -.0732263 > >> 34.3665 > >> > > >> > . sum centyear2 > >> > > >> > Variable | Obs Mean Std. Dev. > Min > >> Max > >> > > -------------+-------------------------------------------------------- > >> > centyear2 | 6192 0 1.000024 > -1.626886 > >> 1.626886 > >> > > >> > . sum centforestarea2 > >> > > >> > Variable | Obs Mean Std. Dev. > Min > >> Max > >> > > -------------+-------------------------------------------------------- > >> > centfores~a2 | 6192 -.0043667 1.00682 -2.396995 > >> 2.746216 > >> > > >> > . sum centgdp2 > >> > > >> > Variable | Obs Mean Std. Dev. > Min > >> Max > >> > > -------------+-------------------------------------------------------- > >> > centgdp2 | 6192 -.0835699 .8318088 -.3333735 > >> 5.257175 > >> > > >> > . sum centlandarea2 > >> > > >> > Variable | Obs Mean Std. Dev. > Min > >> Max > >> > > -------------+-------------------------------------------------------- > >> > centlandar~2 | 6192 -.0336882 .9528875 -.6987395 > >> 2.490177 > >> > > >> > . sum centpopulation2 > >> > > >> > Variable | Obs Mean Std. Dev. > Min > >> Max > >> > > -------------+-------------------------------------------------------- > >> > centpopul~n2 | 6192 -.0018452 1.069818 -.6711841 > >> 8.741787 > >> > > >> > At last a short extract of the dataset, with the original quantity, > in > >> order to see the structure: > >> > > >> > quantity country genus centyear2 centagriculturalland2 > >> > 0 USA Tursiops 1.19305 .6379885 > >> > 0 USA Tursiops 1.409968 .6239355 > >> > 6.08 USA Tursiops 1.626886 .6126072 > >> > 40.29 USA Ursus -1.626886 .7022238 > >> > 65.375 USA Ursus -1.409968 .7022238 > >> > 140.255 USA Ursus -1.19305 .6926397 > >> > > >> > In total 40 countries, 213 genus and 21 independendt variables over a > >> period of 16 years with 6192 observations. As there is no hirarchical > >> structure, there are no different levels. There is one level for the > quantity > >> exported and two randome effects at this level, the country and genus. > >> > After having transformed the quantity to z-scores, would you still > >> recommend dividing it by 100k? > >> > > >> > Thank you > >> > > >> > Lukas > >> > > >> > > >> > -------- Original-Nachricht -------- > >> >> Datum: Thu, 23 Jun 2011 14:41:44 -0400 > >> >> Von: Joerg Luedicke <joerg.luedicke@gmail.com> > >> >> An: statalist@hsphsun2.harvard.edu > >> >> Betreff: Re: st: xtmixed with nonrtolerance. What happens? > >> > > >> >> k stands for 1000 (as in kb=1000 bytes, for instance). What are your > >> >> Level 1 observations (i.e., the 6192)? If only 72 bears were > exported > >> >> from the US in a given year then figures in the ballpark of hundreds > >> >> of thousands appear fairly high to me? > >> >> > >> >> J. > >> >> > >> >> On Thu, Jun 23, 2011 at 2:14 PM, "Lukas Bösch" <L.Boesch@gmx.de> > >> wrote: > >> >> > In my opinion the scales dont differ wildly. > >> >> > I am not a statistician though, so maybe you have a different > >> opinion. > >> >> > > >> >> > > >> >> > . sum centgdp2 > >> >> > > >> >> > Variable | Obs Mean Std. Dev. > >> Min > >> >> Max > >> >> > > >> -------------+-------------------------------------------------------- > >> >> > centgdp2 | 6192 -.0835699 .8318088 > -.3333735 > >> >> 5.257175 > >> >> > > >> >> > . sum centlandarea2 > >> >> > > >> >> > Variable | Obs Mean Std. Dev. > >> Min > >> >> Max > >> >> > > >> -------------+-------------------------------------------------------- > >> >> > centlandar~2 | 6192 -.0336882 .9528875 -.6987395 > >> >> 2.490177 > >> >> > > >> >> > . sum centpopulation2 > >> >> > > >> >> > Variable | Obs Mean Std. Dev. > >> Min > >> >> Max > >> >> > > >> -------------+-------------------------------------------------------- > >> >> > centpopul~n2 | 6192 -.0018452 1.069818 -.6711841 > >> >> 8.741787 > >> >> > > >> >> > . sum centyear2 > >> >> > > >> >> > Variable | Obs Mean Std. Dev. > >> Min > >> >> Max > >> >> > > >> -------------+-------------------------------------------------------- > >> >> > centyear2 | 6192 0 1.000024 > >> -1.626886 > >> >> 1.626886 > >> >> > > >> >> > . sum centforestarea2 > >> >> > > >> >> > Variable | Obs Mean Std. Dev. > >> Min > >> >> Max > >> >> > > >> -------------+-------------------------------------------------------- > >> >> > centfores~a2 | 6192 -.0043667 1.00682 -2.396995 > >> >> 2.746216 > >> >> > > >> >> > The dependent variable is export. The export of wild animal and > plant > >> >> products from one country to the rest of the world. For example: US > >> export of > >> >> Bears in 1992: 72. > >> >> > Because I cannot sum up the export of different species to one > export > >> >> figure, obviously bears and pearls are not the same, i have to deal > >> with > >> >> those mixed models. Socioeconomic factors are set as fixed effects > and > >> the > >> >> genus and countries as the variable effects. > >> >> > As one species can be exported by different countries, the data is > >> not > >> >> hierarchic and country and genus are cross-classified. Or i think > this > >> is > >> >> what it means. Two random effects at the same level for all > >> observations. > >> >> Joerge, can you explain what you mean with dividing by 100k? What > does > >> the k > >> >> stand for? > >> >> > > >> >> > Thank you > >> >> > > >> >> > Lukas > >> >> > > >> >> > mixed modells-------- Original-Nachricht -------- > >> >> >> Datum: Thu, 23 Jun 2011 09:47:55 -0400 > >> >> >> Von: Joerg Luedicke <joerg.luedicke@gmail.com> > >> >> >> An: statalist@hsphsun2.harvard.edu > >> >> >> Betreff: Re: st: xtmixed with nonrtolerance. What happens? > >> >> > > >> >> >> Your model did not converge using the default convergence > criteria > >> and > >> >> >> with -nonrtolerance- you just turned off that default criteria > >> >> >> (though, I do not know what criteria is used instead?). However, > you > >> >> >> should be very cautious with regard to the results. > >> >> >> > >> >> >> What is your dependent variable? From your output I gather that > its > >> >> >> predicted mean is roughly 900k at average values of your > covariates. > >> >> >> Maybe you should transform your dependent variable and fit the > model > >> >> >> again (e.g., dividing it by 100k). > >> >> >> > >> >> >> A question in regards to your random effects: are -country- and > >> >> >> -genus- cross-classified? > >> >> >> > >> >> >> J. > >> >> >> > >> >> >> On Thu, Jun 23, 2011 at 6:21 AM, "Lukas Bösch" <L.Boesch@gmx.de> > >> >> wrote: > >> >> >> > I transformed the data to z-scores (score-mean/stdeviation) > before > >> >> doing > >> >> >> the regression. > >> >> >> > What do you mean with differing scales? I have either percents, > >> for > >> >> >> example % forest area, or absolute figures, for example land > area, > >> in > >> >> my > >> >> >> dataset, but they are all transformed and should therefore be > >> uniform. > >> >> >> > What about nonrtolerance? > >> >> >> > > >> >> >> > Thank you > >> >> >> > > >> >> >> > Lukas > >> >> >> > > >> >> >> > -------- Original-Nachricht -------- > >> >> >> >> Datum: Wed, 22 Jun 2011 18:48:22 -0400 > >> >> >> >> Von: Stas Kolenikov <skolenik@gmail.com> > >> >> >> >> An: statalist@hsphsun2.harvard.edu > >> >> >> >> Betreff: Re: st: xtmixed with nonrtolerance. What happens? > >> >> >> > > >> >> >> >> It looks like you have data with wildly differing scales. I > >> >> understand > >> >> >> >> that you need to interpret the results in the original scales, > >> but > >> >> >> >> maybe you could rescale your variables so that all of your > >> >> >> >> coefficients would be about 1. Whether that will help > convergence > >> is > >> >> >> >> anybody's telling, of course, but usually differences in the > >> scales > >> >> >> >> (and hence coefficients) of the order of 1e3-1e4 are > detrimental > >> to > >> >> >> >> numeric convergence. > >> >> >> >> > >> >> >> >> On Wed, Jun 22, 2011 at 4:33 PM, "Lukas Bösch" > <L.Boesch@gmx.de> > >> >> >> wrote: > >> >> >> >> > Dear Statalist community. > >> >> >> >> > > >> >> >> >> > I am using Stata 10.0 and doing a mixed model analysis of > >> export > >> >> >> data. > >> >> >> >> > After trying different options and always having trouble to > get > >> a > >> >> >> >> propper output i finally found a way to get to my results. I > >> however > >> >> >> could not > >> >> >> >> find any information about why it works and if it is allright. > >> But > >> >> let > >> >> >> us > >> >> >> >> first start with the problem: > >> >> >> >> > > >> >> >> >> > 1) This is the command i enter and the output stata creates: > >> >> >> >> > > >> >> >> >> > xtmixed quantity year centforestarea2 centgdp2 centlandarea2 > >> >> >> >> centpopulation2 || _all: R.country || _all: R.genus > >> >> >> >> > > >> >> >> >> > Performing EM optimization: > >> >> >> >> > > >> >> >> >> > Performing gradient-based optimization: > >> >> >> >> > > >> >> >> >> > Iteration 0: log restricted-likelihood = -77051.164 > >> >> >> >> > Iteration 1: log restricted-likelihood = -77046.704 > >> >> >> >> > Iteration 2: log restricted-likelihood = -77046.565 > >> >> >> >> > Iteration 3: log restricted-likelihood = -77046.5 > >> >> >> >> > Iteration 4: log restricted-likelihood = -77046.468 > >> (backed > >> >> up) > >> >> >> >> > Iteration 5: log restricted-likelihood = -77046.46 > >> (backed > >> >> up) > >> >> >> >> > Iteration 6: log restricted-likelihood = -77046.456 > >> (backed > >> >> up) > >> >> >> >> > Iteration 7: log restricted-likelihood = -77046.454 > >> (backed > >> >> up) > >> >> >> >> > numerical derivatives are approximate > >> >> >> >> > nearby values are missing > >> >> >> >> > Iteration 8: log restricted-likelihood = -77046.453 > >> (backed > >> >> up) > >> >> >> >> > numerical derivatives are approximate > >> >> >> >> > nearby values are missing > >> >> >> >> > Hessian has become unstable or asymmetric > >> >> >> >> > > >> >> >> >> > Mixed-effects REML regression > Number > >> of > >> >> >> obs > >> >> >> >> = 6192 > >> >> >> >> > Group variable: _all > >> >> Number > >> >> >> of > >> >> >> >> groups = 1 > >> >> >> >> > > >> >> >> >> > > >> >> >> >> Obs per group: min = 6192 > >> >> >> >> > > >> >> >> >> avg = 6192.0 > >> >> >> >> > > >> >> >> >> max = 6192 > >> >> >> >> > > >> >> >> >> Wald chi2(5) = 9.26 > >> >> >> >> > Log restricted-likelihood = -77051.164 Prob > > >> chi2 > >> >> >> >> = 0.0991 > >> >> >> >> > quantity | Coef. Std. Err. z > P>|z| > >> >> >> >> [95% Conf. Interval] > >> >> >> >> > year | -429.7599 215.8898 -1.99 > 0.047 > >> >> >> >> -852.8961 -6.623654 > >> >> >> >> > centfores~a2 | -9875.264 6631.861 -1.49 0.136 > >> >> >> >> -22873.47 3122.945 > >> >> >> >> > centgdp2 | -2024.629 4138.469 -0.49 0.625 > >> >> >> >> -10135.88 6086.621 > >> >> >> >> > centlandar~2 | -52889.76 63817.96 -0.83 0.407 > >> >> >> >> -177970.7 72191.13 > >> >> >> >> > centpopul~n2 | 22296.98 10234.72 2.18 0.029 > >> >> >> >> 2237.304 42356.66 > >> >> >> >> > _cons | 895402.2 433369.4 2.07 0.039 > >> >> >> >> 46013.74 1744791 > >> >> >> >> > > >> >> >> >> > Random-effects Parameters | Estimate Std. Err. > >> >> [95% > >> >> >> >> Conf. Interval] > >> >> >> >> > > >> >> >> >> > _all: Identity | > >> >> >> >> > sd(R.country) | 313329.2 > >> . > >> >> >> >> > _all: Identity | > >> >> >> >> > sd(R.genus) | 6757.304 > >> . > >> >> >> >> > sd(Residual) | 60169.26 > >> . > >> >> >> >> > LR test vs. linear regression: chi2(2) = 7810.42 > >> >> Prob > > >> >> >> >> chi2 = 0.0000 > >> >> >> >> > > >> >> >> >> > Note: LR test is conservative and provided only for > reference. > >> >> >> >> > Warning: convergence not achieved; estimates are based on > >> iterated > >> >> EM > >> >> >> >> > > >> >> >> >> > Obviously Stata has a problem and can't calculate the > standard > >> >> errors > >> >> >> of > >> >> >> >> the random factors. > >> >> >> >> > > >> >> >> >> > 2) With the option nonrtolerance it works however: > >> >> >> >> > > >> >> >> >> > xtmixed quantity year centforestarea2 centgdp2 centlandarea2 > >> >> >> >> centpopulation2 || _all: R.country || _all: R.genus, > >> nonrtolerance > >> >> >> >> > > >> >> >> >> > Performing EM optimization: > >> >> >> >> > > >> >> >> >> > Performing gradient-based optimization: > >> >> >> >> > > >> >> >> >> > Iteration 0: log restricted-likelihood = -77051.164 > >> >> >> >> > Iteration 1: log restricted-likelihood = -77046.704 > >> >> >> >> > Iteration 2: log restricted-likelihood = -77046.565 > >> >> >> >> > Iteration 3: log restricted-likelihood = -77046.5 > >> >> >> >> > Iteration 4: log restricted-likelihood = -77046.468 > >> (backed > >> >> up) > >> >> >> >> > Iteration 5: log restricted-likelihood = -77046.46 > >> (backed > >> >> up) > >> >> >> >> > Iteration 6: log restricted-likelihood = -77046.456 > >> (backed > >> >> up) > >> >> >> >> > > >> >> >> >> > Computing standard errors: > >> >> >> >> > > >> >> >> >> > Mixed-effects REML regression > Number > >> of > >> >> >> obs > >> >> >> >> = 6192 > >> >> >> >> > Group variable: _all > >> >> Number > >> >> >> of > >> >> >> >> groups = 1 > >> >> >> >> > > >> >> >> >> > > >> >> >> >> Obs per group: min = 6192 > >> >> >> >> > > >> >> >> >> avg = 6192.0 > >> >> >> >> > > >> >> >> >> max = 6192 > >> >> >> >> > > >> >> >> >> > > >> >> >> >> > > >> >> >> >> Wald chi2(5) = 9.22 > >> >> >> >> > Log restricted-likelihood = -77046.456 Prob > > >> chi2 > >> >> >> >> = 0.1008 > >> >> >> >> > quantity | Coef. Std. Err. z > P>|z| > >> >> >> >> [95% Conf. Interval] > >> >> >> >> > year | -429.7645 216.4073 -1.99 > 0.047 > >> >> >> >> -853.915 -5.614053 > >> >> >> >> > centfores~a2 | -9885.307 6647.52 -1.49 0.137 > >> >> >> >> -22914.21 3143.592 > >> >> >> >> > centgdp2 | -2021.312 4148.464 -0.49 0.626 > >> >> >> >> -10152.15 6109.527 > >> >> >> >> > centlandar~2 | -52859.75 63778.66 -0.83 0.407 > >> >> >> >> -177863.6 72144.12 > >> >> >> >> > centpopul~n2 | 22276.96 10257.46 2.17 0.030 > >> >> >> >> 2172.715 42381.2 > >> >> >> >> > _cons | 895338.1 434389.3 2.06 0.039 > >> >> >> >> 43950.68 1746726 > >> >> >> >> > > >> >> >> >> > Random-effects Parameters | Estimate Std. Err. > >> >> [95% > >> >> >> >> Conf. Interval] > >> >> >> >> > _all: Identity | > >> >> >> >> > sd(R.country) | 313133.2 > 36075.6 > >> >> >> >> 249840.9 392459.4 > >> >> >> >> > _all: Identity | > >> >> >> >> > sd(R.genus) | 3440.288 > 1355.694 > >> >> >> >> 1589.157 7447.712 > >> >> >> >> > sd(Residual) | 60315.87 > 545.9681 > >> >> >> >> 59255.23 61395.5 > >> >> >> >> > LR test vs. linear regression: chi2(2) = 7819.83 > >> >> Prob > > >> >> >> >> chi2 = 0.0000 > >> >> >> >> > Note: LR test is conservative and provided only for > reference. > >> >> >> >> > > >> >> >> >> > Can someone explain to me why it works with nonrtolerance > and > >> tell > >> >> me > >> >> >> if > >> >> >> >> these outputs are as reliable as if they were created without > >> >> >> >> nonrtolerance. I searched in the stata help and on stata.com > but > >> >> could > >> >> >> not find more > >> >> >> >> information about this. > >> >> >> >> > > >> >> >> >> > Kind regards > >> >> >> >> > > >> >> >> >> > Lukas > >> >> >> >> > > >> >> >> >> > -- > >> >> >> >> > NEU: FreePhone - kostenlos mobil telefonieren! > >> >> >> >> > Jetzt informieren: http://www.gmx.net/de/go/freephone > >> >> >> >> > * > >> >> >> >> > * For searches and help try: > >> >> >> >> > * http://www.stata.com/help.cgi?search > >> >> >> >> > * http://www.stata.com/support/statalist/faq > >> >> >> >> > * http://www.ats.ucla.edu/stat/stata/ > >> >> >> >> > > >> >> >> >> > >> >> >> >> > >> >> >> >> > >> >> >> >> -- > >> >> >> >> Stas Kolenikov, also found at http://stas.kolenikov.name > >> >> >> >> Small print: I use this email account for mailing lists only. > >> >> >> >> > >> >> >> >> * > >> >> >> >> * For searches and help try: > >> >> >> >> * http://www.stata.com/help.cgi?search > >> >> >> >> * http://www.stata.com/support/statalist/faq > >> >> >> >> * http://www.ats.ucla.edu/stat/stata/ > >> >> >> > > >> >> >> > -- > >> >> >> > NEU: FreePhone - kostenlos mobil telefonieren! > >> >> >> > Jetzt informieren: http://www.gmx.net/de/go/freephone > >> >> >> > * > >> >> >> > * For searches and help try: > >> >> >> > * http://www.stata.com/help.cgi?search > >> >> >> > * http://www.stata.com/support/statalist/faq > >> >> >> > * http://www.ats.ucla.edu/stat/stata/ > >> >> >> > > >> >> >> > >> >> >> * > >> >> >> * For searches and help try: > >> >> >> * http://www.stata.com/help.cgi?search > >> >> >> * http://www.stata.com/support/statalist/faq > >> >> >> * http://www.ats.ucla.edu/stat/stata/ > >> >> > > >> >> > -- > >> >> > Empfehlen Sie GMX DSL Ihren Freunden und Bekannten und wir > >> >> > belohnen Sie mit bis zu 50,- Euro! > >> https://freundschaftswerbung.gmx.de > >> >> > * > >> >> > * For searches and help try: > >> >> > * http://www.stata.com/help.cgi?search > >> >> > * http://www.stata.com/support/statalist/faq > >> >> > * http://www.ats.ucla.edu/stat/stata/ > >> >> > > >> >> > >> >> * > >> >> * For searches and help try: > >> >> * http://www.stata.com/help.cgi?search > >> >> * http://www.stata.com/support/statalist/faq > >> >> * http://www.ats.ucla.edu/stat/stata/ > >> > > >> > -- > >> > NEU: FreePhone - kostenlos mobil telefonieren! > >> > Jetzt informieren: http://www.gmx.net/de/go/freephone > >> > * > >> > * For searches and help try: > >> > * http://www.stata.com/help.cgi?search > >> > * http://www.stata.com/support/statalist/faq > >> > * http://www.ats.ucla.edu/stat/stata/ > >> > > >> > >> * > >> * For searches and help try: > >> * http://www.stata.com/help.cgi?search > >> * http://www.stata.com/support/statalist/faq > >> * http://www.ats.ucla.edu/stat/stata/ > > > > -- > > NEU: FreePhone - kostenlos mobil telefonieren! > > Jetzt informieren: http://www.gmx.net/de/go/freephone > > * > > * For searches and help try: > > * http://www.stata.com/help.cgi?search > > * http://www.stata.com/support/statalist/faq > > * http://www.ats.ucla.edu/stat/stata/ > > > > * > * For searches and help try: > * http://www.stata.com/help.cgi?search > * http://www.stata.com/support/statalist/faq > * http://www.ats.ucla.edu/stat/stata/ -- Empfehlen Sie GMX DSL Ihren Freunden und Bekannten und wir belohnen Sie mit bis zu 50,- Euro! https://freundschaftswerbung.gmx.de * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/

**Follow-Ups**:**Re: st: xtmixed with nonrtolerance. What happens?***From:*Nick Cox <njcoxstata@gmail.com>

**References**:**st: xtmixed with nonrtolerance. What happens?***From:*"Lukas Bösch" <L.Boesch@gmx.de>

**Re: st: xtmixed with nonrtolerance. What happens?***From:*Stas Kolenikov <skolenik@gmail.com>

**Re: st: xtmixed with nonrtolerance. What happens?***From:*"Lukas Bösch" <L.Boesch@gmx.de>

**Re: st: xtmixed with nonrtolerance. What happens?***From:*Joerg Luedicke <joerg.luedicke@gmail.com>

**Re: st: xtmixed with nonrtolerance. What happens?***From:*"Lukas Bösch" <L.Boesch@gmx.de>

**Re: st: xtmixed with nonrtolerance. What happens?***From:*Joerg Luedicke <joerg.luedicke@gmail.com>

**Re: st: xtmixed with nonrtolerance. What happens?***From:*"Lukas Bösch" <L.Boesch@gmx.de>

**Re: st: xtmixed with nonrtolerance. What happens?***From:*Joerg Luedicke <joerg.luedicke@gmail.com>

**Re: st: xtmixed with nonrtolerance. What happens?***From:*"Lukas Bösch" <L.Boesch@gmx.de>

**Re: st: xtmixed with nonrtolerance. What happens?***From:*Joerg Luedicke <joerg.luedicke@gmail.com>

- Prev by Date:
**st: dynamic panel - bootstrapped Sargan and exclusion tests** - Next by Date:
**st: Estimating a mixed logit model with mixlogit** - Previous by thread:
**Re: st: xtmixed with nonrtolerance. What happens?** - Next by thread:
**Re: st: xtmixed with nonrtolerance. What happens?** - Index(es):