as.mlm: Refit Constrained Ordination as a Multiple Response Linear...

Description Usage Arguments Details Value Note Author(s) See Also Examples

Description

Functions refit results of constrained ordination (cca, rda, capscale) as a multiple response linear model (lm). This allows finding influence statistics (influence.measures). This also allows deriving several other statistics, but most of these are biased and misleading, since refitting ignores a major component of variation in constrained ordination.

Usage

1
as.mlm(x)

Arguments

x

Constrained ordination result.

Details

Popular algorithm for constrained ordination is based on iteration with regression where weighted averages of sites are used as dependent variables and constraints as independent variables. Statistics of linear regression are a natural by-product in this algorithm. Constrained ordination in vegan uses different algorithm, but to obtain linear regression statistics you can refit an ordination result as a multiple response linear model (lm). This regression ignores residual unconstrained variation in the data, and therefore estimates of standard error are strongly biased and much too low. You can get statistics like t-values of coefficients, but you should not use these because of this bias. Some useful information you can get with refitted models are statistics for detecting influential observations (influence.measures including cooks.distance, hatvalues).

Value

Function returns an object of multiple response linear model of class "mlm" documented with lm.

Note

You can use these functions to find t-values of coefficients using summary.mlm, but you should not do this because the method ignores unconstrained residual variation. You also can find several other statistics for (multiple response) linear models with similar bias. This bias is not a unique feature in vegan implementation, but also applies to implementations in other software.

Some statistics of linear models can be found without using these functions: coef.cca gives the regression coefficients, spenvcor the species-environment correlation, intersetcor the interset correlation, vif.cca the variance inflation factors.

Author(s)

Jari Oksanen

See Also

cca, rda, capscale, cca.object, lm, summary.mlm, influence.measures.

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
data(varespec)
data(varechem)
mod <- cca(varespec ~ Al + P + K, data=varechem)
lmod <- as.mlm(mod)
## Coefficients
lmod
coef(mod)
## Influential observations
influence.measures(lmod)
plot(mod, type = "n")
points(mod, cex = 10*hatvalues(lmod), pch=16, xpd = TRUE)
text(mod, display = "bp", col = "blue") 

Example output

Loading required package: permute
Loading required package: lattice
This is vegan 2.5-4
Warning message:
'as.mlm' is deprecated.
Use 'see ?hatvalues.cca for new alternatives' instead.
See help("Deprecated") 

Call:
lm(formula = WA ~ ., data = X, weights = w)

Coefficients:
             CCA1        CCA2        CCA3      
(Intercept)  -1.364e-16   1.298e-16   2.909e-16
Al            7.479e-03  -1.884e-03   3.381e-03
P            -6.491e-03  -1.022e-01  -2.231e-02
K            -6.756e-03   1.534e-02   1.707e-02

           CCA1         CCA2         CCA3
Al  0.007478556 -0.001883637  0.003380774
P  -0.006491081 -0.102189737 -0.022306682
K  -0.006755568  0.015343662  0.017067351
Influence measures of
	 lm(formula = WA ~ ., data = X, weights = w) :

     dfb.1_    dfb.Al    dfb.P    dfb.K     CCA1    CCA2      CCA3   cov.r
18  0.29810 -0.255305  0.00992 -0.06408  0.40781  0.2688 -0.176741 0.00805
15 -0.18825  0.101295  0.14064 -0.11951 -0.25218 -0.1425  0.010516 0.01045
24 -0.28878 -0.003523 -0.45031  0.21238 -0.57120 -0.2072 -0.308800 0.00960
27 -0.06494  0.074353 -0.01770 -0.03647 -0.12401 -0.2040  0.027117 0.01426
23  0.12010 -0.118311  0.07005 -0.02583  0.18926  0.3833 -0.289598 0.01201
19  0.01897 -0.007462 -0.01180  0.01090  0.02342 -0.3989 -0.000717 0.01232
22 -0.13652  0.155052  0.15252 -0.13450 -0.25070  0.4673  0.206913 0.01287
16 -0.16882  0.109056  0.18167 -0.10065 -0.27346  0.2903  0.250839 0.01121
28 -0.31694  0.344951 -0.35723 -0.05623 -0.77163  0.0845  0.707320 0.01164
13  0.37299  0.421402 -1.15819  1.42065  1.58279  0.4848  0.327464 0.04124
14 -0.05322  0.024374  0.02536 -0.01172 -0.06449  0.2666  0.095880 0.01212
20 -0.00544  0.000749 -0.00561  0.00206 -0.00857  0.3807 -0.246898 0.01254
25 -0.16251  0.169698 -0.11245  0.09507 -0.26813 -0.2082 -0.090535 0.01155
7  -0.31124 -0.408929  0.16214  0.16126 -0.68366  0.6756 -0.969451 0.01028
5   0.37506 -0.287253 -0.09355 -0.36540  0.84290  0.4232 -0.374102 0.00831
6   0.01668  0.003281 -0.00173 -0.01594  0.03142  0.4182  0.243918 0.01398
3  -0.21235 -0.312389 -0.08137  0.26748 -0.48208 -0.3178  0.615865 0.01288
4  -0.02435 -0.059559 -0.02785  0.00891 -0.07227  0.8919  0.142503 0.01692
2   0.01444  0.021023  0.00430 -0.00385  0.02565 -0.5189 -0.200173 0.01386
9   0.09453  0.076663  0.09866 -0.11215  0.16179 -0.4431  0.820408 0.01333
12  0.32483 -0.100807 -0.21465  0.04778  0.43890 -1.0115  0.414212 0.00874
10  0.75838  0.156020  0.69648 -0.13570  1.20348 -0.3381  0.000769 0.00250
11  0.01686  0.015066  0.00709  0.00108  0.02587  0.2486 -0.407255 0.01315
21 -0.16486  0.151470  0.15801 -0.02979 -0.30670 -0.2293 -0.385998 0.01213
     CCA1.1   CCA2.1   CCA3.1    hat inf
18 1.22e-02 0.005316 2.30e-03 0.0690    
15 5.00e-03 0.001595 8.69e-06 0.0667    
24 2.45e-02 0.003224 7.16e-03 0.1525    
27 1.26e-03 0.003413 6.03e-05 0.1894    
23 2.89e-03 0.011865 6.77e-03 0.0929    
19 4.51e-05 0.013083 4.23e-08 0.0512    
22 5.08e-03 0.017638 3.46e-03 0.1531    
16 5.93e-03 0.006683 4.99e-03 0.0961    
28 4.52e-02 0.000542 3.80e-02 0.2714    
13 1.98e-01 0.018551 8.47e-03 0.7589   *
14 3.41e-04 0.005822 7.53e-04 0.0496    
20 6.04e-06 0.011917 5.01e-03 0.0658    
25 5.73e-03 0.003455 6.53e-04 0.1059    
7  3.51e-02 0.034299 7.06e-02 0.2063    
5  5.07e-02 0.012793 1.00e-02 0.1980    
6  8.12e-05 0.014379 4.89e-03 0.1628    
3  1.84e-02 0.007978 3.00e-02 0.2274    
4  4.29e-04 0.065381 1.67e-03 0.3092    
2  5.41e-05 0.022141 3.30e-03 0.1556    
9  2.14e-03 0.016020 5.49e-02 0.1486    
12 1.44e-02 0.076367 1.28e-02 0.0905   *
10 7.83e-02 0.006177 3.19e-08 0.1275   *
11 5.50e-05 0.005081 1.36e-02 0.1098    
21 7.51e-03 0.004198 1.19e-02 0.1420    

vegan documentation built on May 2, 2019, 5:51 p.m.