[R] Problem in lme longitudinal model

[Bernardo Rangel Tura]

Hi R-masters!

I trying model Heart disease mortality in my country with a lme model like 
this:

m1.lme<-lme(log(rdeath)~age*year,random=~age|year,data=dados)

where: rdeath is rate of mortality per 100000 person per age and year
            age: age of death (22 27 32 37 42 47 52 57 62 67 72 77 82)
            year: year of death (1980:2002)

I don´t have problem to fit  the model, but in residual analysis I have one 
problem.

If i type acf(m1.lme$residuals) the graph show 4 plot with a sin curve (a 
example in attach) and I get this variogram:

 > Variogram(m1.lme)
       variog dist n.pairs
1  0.3939065    1     276
2  0.6486452    2     253
3  0.9679870    3     230
4  1.2765094    4     207
5  1.4158147    5     184
6  1.4264685    6     161
7  1.4048608    7     138
8  1.1902613    8     115
9  0.9619330    9      92
10 0.7037427   10      69
11 0.8591257   11      46
12 1.2215657   12      23

Well
I need help for solution this problem....
Somebody can help me?


Thanks in advance

Bernardo Rangel Tura, MD, MSc
National Institute of Cardiology Laranjeiras
Rio de Janeiro Brazil 
-------------- next part --------------


No virus found in this outgoing message.
Checked by AVG Anti-Virus.