Model selection for restored prairie data

Obs site_id age linear ph organic TE PE TA PA exotic bc y
1 3 5 1.29 7.33 12.28 12.25 31.77 9.94 35.61 24.05 0.982 0.982
2 4 6 1.39 7.93 8.54 8.69 45.50 8.89 36.80 19.87 0.898 0.898
3 5 14 1.21 7.90 7.11 9.31 31.83 9.11 35.52 3.68 0.791 0.791
4 7 5 1.24 8.03 6.28 9.42 28.26 8.00 36.48 10.25 1.000 1.000
5 8 3 1.28 7.67 5.69 6.61 43.13 8.00 36.48 18.00 0.998 0.998
6 9 10 1.12 7.97 7.92 8.97 41.19 8.00 36.48 4.50 0.712 0.712
7 10 8 1.04 8.10 11.04 7.19 28.99 7.22 29.60 0.90 0.623 0.623
8 13 2 1.22 7.53 9.43 7.06 33.20 8.94 30.46 17.70 0.892 0.892
9 14 6 1.43 7.80 7.81 7.42 37.36 8.94 30.46 19.70 0.841 0.841
10 15 8 1.10 6.70 11.44 7.72 38.73 7.94 33.94 12.97 0.646 0.646
11 22 24 1.33 7.70 8.94 8.11 46.30 9.06 34.67 13.65 0.994 0.994
12 23 6 1.06 6.47 4.31 9.81 27.33 7.89 36.66 9.20 0.676 0.676
13 30 4 1.47 7.80 15.56 11.89 29.17 9.72 36.09 8.33 0.994 0.994
14 31 5 1.29 7.60 10.03 12.25 31.77 9.94 35.61 18.08 0.894 0.894
15 32 5 1.36 7.53 9.07 12.25 31.77 9.94 35.61 23.85 0.953 0.953
16 33 7 1.06 6.77 9.46 8.22 43.67 8.28 35.84 0.20 0.474 0.474
17 34 9 1.02 6.53 10.21 7.81 41.78 8.28 35.84 5.67 0.503 0.503
18 35 6 1.06 6.47 10.47 8.53 38.07 8.28 35.84 10.14 0.684 0.684
19 36 9 1.03 6.33 10.29 7.81 41.78 8.28 35.84 6.33 0.632 0.632
20 39 14 1.02 7.30 9.39 10.50 30.26 9.72 36.27 3.57 0.589 0.589
21 40 15 1.12 7.63 13.47 10.72 29.43 10.50 36.02 21.71 1.000 1.000
22 41 6 1.02 7.57 5.98 8.69 45.50 8.67 36.47 0.09 0.315 0.315
23 42 4 1.28 6.77 3.78 11.89 29.17 9.72 36.09 11.35 1.000 1.000
24 43 4 1.34 7.93 9.06 11.89 29.17 9.72 36.09 19.10 1.000 1.000
25 44 6 1.42 7.57 9.12 8.69 45.50 8.89 36.80 23.30 0.924 0.924
26 45 9 1.11 6.20 8.62 7.94 42.92 8.56 35.24 9.08 0.660 0.660
27 56 5 1.04 6.97 10.81 9.42 29.81 8.72 38.35 16.75 0.967 0.967
28 65 9 1.22 7.37 17.36 8.06 41.38 9.06 34.23 18.00 0.808 0.808
29 74 12 1.02 7.27 12.30 9.33 32.16 7.89 36.66 3.17 0.739 0.739
30 75 6 1.08 7.30 8.03 9.81 27.33 7.89 36.66 2.22 0.634 0.634
31 77 9 1.32 7.93 8.03 5.94 39.49 7.17 38.30 16.43 0.965 0.965
32 80 10 1.27 8.13 5.82 7.56 41.58 8.17 34.42 18.28 0.959 0.959
33 89 11 1.02 7.40 7.26 9.19 32.51 8.39 35.63 13.00 0.735 0.735
34 90 3 1.18 7.57 10.06 7.61 41.01 8.39 34.67 20.60 0.926 0.926
35 91 6 1.02 7.00 10.90 9.03 24.98 7.72 28.71 10.30 0.680 0.680
36 92 9 1.05 7.07 10.37 7.81 29.42 8.94 30.46 22.72 0.865 0.865
37 96 14 1.15 7.33 7.11 10.67 32.95 9.00 37.38 28.40 0.955 0.955
38 97 5 1.00 6.73 6.22 9.64 31.49 9.50 35.36 7.53 0.916 0.916
39 98 3 1.36 8.10 7.82 9.28 46.10 10.39 37.25 21.40 0.937 0.937
40 100 17 1.30 7.87 10.42 9.17 33.58 10.72 37.59 25.03 0.970 0.970
41 101 3 1.26 7.93 8.64 8.56 40.66 10.72 37.59 15.37 0.772 0.772
42 102 11 1.25 8.10 3.61 11.69 40.36 10.00 36.28 9.57 0.972 0.972
43 105 13 1.05 7.93 14.54 8.53 35.57 8.00 36.48 13.77 0.718 0.718
44 106 10 1.02 6.97 8.19 7.64 47.79 8.00 36.48 14.25 0.624 0.624
Model selection for restored prairie data; Plot of y by age y 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 age 0 10 20 30 Model selection for restored prairie data
Model selection for restored prairie data; Plot of y by linear y 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 linear 1.00 1.05 1.10 1.15 1.20 1.25 1.30 1.35 1.40 1.45 1.50 Model selection for restored prairie data
Model selection for restored prairie data; Plot of y by ph y 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 ph 6.2 6.4 6.6 6.8 7.0 7.2 7.4 7.6 7.8 8.0 8.2 Model selection for restored prairie data
Model selection for restored prairie data; Plot of y by organic y 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 organic 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Model selection for restored prairie data
Model selection for restored prairie data; Plot of y by TE y 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 TE 5 6 7 8 9 10 11 12 13 Model selection for restored prairie data
Model selection for restored prairie data; Plot of y by PE y 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 PE 20 30 40 50 Model selection for restored prairie data
Model selection for restored prairie data; Plot of y by TA y 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 TA 7 8 9 10 11 Model selection for restored prairie data
Model selection for restored prairie data; Plot of y by PA y 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 PA 28 29 30 31 32 33 34 35 36 37 38 39 Model selection for restored prairie data
Model selection for restored prairie data; Plot of y by exotic y 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 exotic 0 10 20 30 Model selection for restored prairie data

Model selection for restored prairie data

The REG Procedure

Model: MODEL1

Dependent Variable: y

Number of Observations Read 44
Number of Observations Used 44
Analysis of Variance
Source DF Sum of
Squares		 Mean
Square		 F Value Pr > F
Model 9 0.95793 0.10644 11.11 <.0001
Error 34 0.32582 0.00958    
Corrected Total 43 1.28375      
Root MSE 0.09789 R-Square 0.7462
Dependent Mean 0.81402 Adj R-Sq 0.6790
Coeff Var 12.02586    
Parameter Estimates
Variable DF Parameter
Estimate		 Standard
Error		 t Value Pr > |t| Standardized
Estimate		 Tolerance Variance
Inflation		 95% Confidence Limits
Intercept 1 -0.06835 0.33572 -0.20 0.8399 0 . 0 -0.75061 0.61392
age 1 0.00358 0.00364 0.98 0.3327 0.09162 0.85869 1.16456 -0.00382 0.01098
linear 1 0.58627 0.17117 3.43 0.0016 0.47598 0.38652 2.58720 0.23841 0.93413
ph 1 0.01789 0.03498 0.51 0.6123 0.05646 0.61263 1.63230 -0.05320 0.08899
organic 1 -0.00646 0.00543 -1.19 0.2425 -0.10727 0.91794 1.08940 -0.01750 0.00458
TE 1 -0.00771 0.01633 -0.47 0.6400 -0.07283 0.31353 3.18951 -0.04090 0.02548
PE 1 -0.01024 0.00333 -3.07 0.0042 -0.38766 0.46845 2.13470 -0.01702 -0.00346
TA 1 0.00021430 0.02467 0.01 0.9931 0.00115 0.42670 2.34355 -0.04992 0.05034
PA 1 0.01084 0.00806 1.35 0.1874 0.13906 0.69869 1.43124 -0.00554 0.02722
exotic 1 0.01060 0.00253 4.19 0.0002 0.46410 0.60956 1.64054 0.00546 0.01573

Model selection for restored prairie data

The REG Procedure

Model: MODEL1

Dependent Variable: y

Model selection for restored prairie data; Panel of fit diagnostics for y. Fit Diagnostics for y 0.679 Adj R-Square 0.7462 R-Square 0.0096 MSE 34 Error DF 10 Parameters 44 Observations Proportion Less 0 1 Residual 0 1 Fit–Mean -0.2 0.0 0.2 -0.28 0.28 Residual 0 10 20 30 Percent 0 10 20 30 40 Observation 0.0 0.1 0.2 0.3 Cook's D 0.4 1.0 Predicted Value 0.4 0.6 0.8 1.0 y -2 -1 0 1 2 Quantile -0.2 0.0 0.2 Residual 0.1 0.3 0.5 Leverage -2 0 2 RStudent 0.5 0.7 0.9 1.1 Predicted Value -2 0 2 RStudent 0.5 0.7 0.9 1.1 Predicted Value -0.2 0.0 0.2 Residual
Model selection for restored prairie data; Panel of scatterplots of residuals by regressors for y. Residual by Regressors for y 25 30 35 40 45 PE 6 8 10 12 TE 5 10 15 organic 6.5 7.0 7.5 8.0 ph 1.0 1.1 1.2 1.3 1.4 linear 5 10 15 20 25 age -0.2 -0.1 0.0 0.1 0.2 Residual -0.2 -0.1 0.0 0.1 0.2 Residual
Model selection for restored prairie data; Panel of scatterplots of residuals by regressors for y. Residual by Regressors for y 0 5 10 15 20 25 exotic 30 32 34 36 38 PA 7 8 9 10 11 TA -0.2 -0.1 0.0 0.1 0.2 Residual

Model selection for restored prairie data

The REG Procedure

Model: MODEL1

Partial Regression Residual Plot

Model selection for restored prairie data; Panel of partial regression scatterplots by regressors for y. Partial Plots for y Partial Regressor Residual Partial Dependent Residual -2 -1 0 1 2 -0.2 0.0 0.2 TE -5 0 5 -0.2 0.0 0.2 organic -1.0 0.5 -0.2 0.0 0.2 ph -0.1 0.2 -0.3 -0.1 0.1 linear -5 5 15 -0.2 0.0 0.2 age -0.1 0.0 0.1 -0.2 0.0 0.2 Intercept
Model selection for restored prairie data; Panel of partial regression scatterplots by regressors for y. Partial Plots for y Partial Regressor Residual Partial Dependent Residual -10 0 10 -0.2 0.0 0.2 exotic -4 -2 0 2 4 -0.2 0.0 0.2 PA -0.5 0.5 1.5 -0.2 0.0 0.2 TA -10 0 10 -0.3 -0.1 0.1 PE

Model selection for restored prairie data

The GLMSELECT Procedure

Data Set WORK.RPDAT
Dependent Variable y
Selection Method Stepwise
Select Criterion AICC
Stop Criterion AICC
Effect Hierarchy Enforced None
Number of Observations Read 44
Number of Observations Used 44
Dimensions
Number of Effects 10
Number of Parameters 10

Model selection for restored prairie data

The GLMSELECT Procedure

Stepwise Selection Summary
Step Effect
Entered		 Effect
Removed		 Number
Effects In		 AICC
* Optimal Value of Criterion
0 Intercept   1 -107.2209
1 exotic   2 -134.3317
2 linear   3 -145.0699
3 PE   4 -153.7590*
Selection stopped at a local minimum of the AICC criterion.
Stop Details
Candidate
For		 Effect Candidate
AICC		   Compare
AICC
Entry PA -153.6592 > -153.7590
Removal PE -145.0699 > -153.7590

Model selection for restored prairie data

The GLMSELECT Procedure

Selected Model

The selected model is the model at the last step (Step 3).

Effects: Intercept linear PE exotic
Analysis of Variance
Source DF Sum of
Squares		 Mean
Square		 F Value
Model 3 0.90599 0.30200 31.98
Error 40 0.37776 0.00944  
Corrected Total 43 1.28375    
Root MSE 0.09718
Dependent Mean 0.81402
R-Square 0.7057
Adj R-Sq 0.6837
AIC -155.33791
AICC -153.75897
SBC -194.20115
Parameter Estimates
Parameter DF Estimate Standard
Error		 t Value
Intercept 1 0.245894 0.147395 1.67
linear 1 0.605915 0.127482 4.75
PE 1 -0.007851 0.002301 -3.41
exotic 1 0.010194 0.002334 4.37