[R-sig-ME] same within group std. error

[peter dalgaard]

On Aug 5, 2012, at 12:39 , Jesus wrote:

> Dear all,
> I have a question regarding mixed effect models and I have not been
> successful in finding the answer in the sites.
> 
> I have my response variable "a" and the explanatory variables "distance",
> "numberAni" and "factor(algorithms)---with  seven different algorithms, I
> also have my random variable "species" (with 5 different species). One of
> the algorithms is set as the intercept in the model.
> 
> I run my model:
> 
> model_1<-(lme(a)~ numberAni*factor(algorithm)+distance*factor(algorithm),
> random=~1|species,data=datamodel)
> 
> An my summary table is as follows. My question is* why* do I get the *same
> Std. Error* for all the algorithms in the first part of the table and then
> other same std. error for the first interactions and then the new one for
> the second interactions? why are the within groups std. errors the same?


Your Subject: suggests that you might not have understood that the column labeled Std.Error is the accuracy of the estimates in the preceding column. This is a function of the estimated variance parameters and of the design -- no. of replications, etc. In a balanced design, which is probably what you have, the factor levels enter interchangeably, so the Std.Error for each level is the same function of the variances. 

> 
> 
>                                    Value  *Std.Error*   DF    t-value p-value
> (Intercept)                     1.0676374 0.28408892 4677   3.758110  0.0002
> NumberAni                       -0.0001166 0.00022606   13  -0.515724  0.6147
> factor(algorithm)A1             -2.8588307 0.14268775 4677 -20.035572  0.0000
> factor(algorithm)A2             -0.3869196 0.14268775 4677  -2.711653  0.0067
> factor(algorithm)A3             -0.6668651 0.14268775 4677  -4.673597  0.0000
> factor(algorithm)A4             -0.6600452 0.14268775 4677  -4.625802  0.0000
> factor(algorithm)A5             -0.3995352 0.14268775 4677  -2.800066  0.0051
> factor(algorithm)A6             -0.8642987 0.14268775 4677  -6.057273  0.0000
> distance                       -0.0000014 0.00000204   13  -0.685703  0.5049
> 
> NumberAni:factor(algorithm)A1   0.0007324 0.00010871 4677   6.737032  0.0000
> 
> NumberAni:factor(algorithm)A2    0.0003483 0.00010871 4677   3.203698  0.0014
> 
> NumberAni:factor(algorithm)A3    0.0004839 0.00010871 4677   4.451614  0.0000
> 
> NumberAni:factor(algorithm)A4    0.0004649 0.00010871 4677   4.277108  0.0000
> 
> NumberAni:factor(algorithm)A5    -0.0000330 0.00010871 4677  -0.303855  0.7613
> 
> NumberAni:factor(algorithm)A6    0.0004812 0.00010871 4677   4.426913  0.0000
> 
> factor(algorithm)A1:distance 0.0000085 0.00000099 4677 8.577618 0.0000
> factor(algorithm)A2:distance 0.0000023 0.00000099 4677 2.309649 0.0210
> factor(algorithm)A3:distance 0.0000024 0.00000099 4677 2.468800 0.0136
> factor(algorithm)A4:distance 0.0000030 0.00000099 4677 3.008320 0.0026
> factor(algorithm)A5:distance -0.0000001 0.00000099 4677 -0.064863 0.9483
> factor(algorithm)A6:distance 0.0000033 0.00000099 4677 3.309633 0.0009
> 
> 
> 
> If someone could explain it to me it would be great. I am sorry if this is
> a question already answered before.
> 
> Regards,
> 
> Jag
> 
> 	[[alternative HTML version deleted]]
> 
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models

-- 
Peter Dalgaard, Professor,
Center for Statistics, Copenhagen Business School
Solbjerg Plads 3, 2000 Frederiksberg, Denmark
Phone: (+45)38153501
Email: pd.mes at cbs.dk  Priv: PDalgd at gmail.com