Hello everyone,
?
Does anybody know whether it's possible to calculate extract major axis
regression residuals on the results of a major axis Model II regression done by
lmodel2?
Many thanks,
Orca
?
Example?:
?
> data1 data1
??? SSTo? SSTp
1? 17.21 20.36
2? 19.61 20.19
3? 18.60 20.08
4? 18.76 19.71
5? 18.24 20.48
6? 19.10 20.56
7? 20.45 20.56
8? 18.68 19.71
9? 16.55 20.48
10 17.33 20.60
11 18.12 18.93
12 18.02 18.65
13 18.43 18.90
14 18.57 18.62
15 17.40 18.79
16 18.91 19.28
17 18.43 18.58
18 17.02 18.40
19 16.96 17.79
20 17.05 17.82
21 16.98 18.07
22 16.86 18.00
23 17.00 17.79
24 16.70 18.18
25 15.93 18.07
26 17.06 17.85
?
>
Reg1<-lmodel2(SSTo~SSTp,data=data1,"interval","interval",99)
> Reg1
?
Model II regression
?
Call: lmodel2(formula = SSTo ~ SSTp, data = data1, range.y
"interval", range.x = "interval", nperm = 99)
?
n = 26?? r = 0.5671202?? r-square = 0.3216254
Parametric P-values:?? 2-tailed = 0.002517785??? 1-tailed = 0.001258892
Angle between the two OLS regression lines = 30.86925 degrees
?
Permutation tests of OLS, MA, RMA slopes: 1-tailed, tail corresponding to sign
A permutation test of r is equivalent to a permutation test of the OLS slope
P-perm for SMA = NA because the SMA slope cannot be tested
?
Regression results
? Method Intercept???? Slope Angle (degrees) P-perm (1-tailed)
1??? OLS? 6.651441 0.5862273??????? 30.38000????????????? 0.01
2???? MA -2.397667 1.0601457??????? 46.67227????????????? 0.01
3??? SMA -1.892541 1.0336913??????? 45.94911??????????????? NA
4?? ?RMA? 3.144884 0.7698721??????? 37.59167????????????? 0.01
?
Confidence intervals
? Method 2.5%-Intercept 97.5%-Intercept 2.5%-Slope 97.5%-Slope
1??? OLS???? -0.2070452?????? 13.509927? 0.2275463?? 0.9449083
2???? MA??? -22.1054387??????? 7.308805? 0.5517999?? 2.0922780
3??? SMA???? -9.8957401??????? 3.801737? 0.7354715?? 1.4528336
4??? RMA???? -7.3480741?????? 11.432676? 0.3358252?? 1.3194076
?
Eigenvalues: 1.751428 0.4828554
?
H statistic used for computing C.I. of MA: 0.09327046
[[alternative HTML version deleted]]