UCLA Academic Technology Services HomeServicesClassesContactJobs
Search

SAS Textbook Examples
Regression Analysis by Example by Chatterjee, Hadi and Price
Chapter 3: Multiple Linear Regression

This page shows how to obtain the results from Chatterjee, Hadi and Price's Chapter 3 using SAS.
Use data in file p054.
data p054;
  input Y X1 X2 X3 X4 X5 X6 ;
cards;
43 51 30 39 61 92 45 
63 64 51 54 63 73 47 
71 70 68 69 76 86 48 
61 63 45 47 54 84 35 
81 78 56 66 71 83 47 
43 55 49 44 54 49 34 
58 67 42 56 66 68 35 
71 75 50 55 70 66 41 
72 82 72 67 71 83 31 
67 61 45 47 62 80 41 
64 53 53 58 58 67 34 
67 60 47 39 59 74 41 
69 62 57 42 55 63 25 
68 83 83 45 59 77 35 
77 77 54 72 79 77 46 
81 90 50 72 60 54 36 
74 85 64 69 79 79 63 
65 60 65 75 55 80 60 
65 70 46 57 75 85 46 
50 58 68 54 64 78 52 
50 40 33 34 43 64 33 
64 61 52 62 66 80 41 
53 66 52 50 63 80 37 
40 37 42 58 50 57 49 
63 54 42 48 66 75 33 
66 77 66 63 88 76 72 
78 75 58 74 80 78 49 
48 57 44 45 51 83 38 
85 85 71 71 77 74 55 
82 82 39 59 64 78 39 
;
run;
Table 3.3, page 54.
proc print data=p054 (obs=4); 
run;

Obs     Y    X1    X2    X3    X4    X5    X6
  1    43    51    30    39    61    92    45
  2    63    64    51    54    63    73    47
  3    71    70    68    69    76    86    48
  4    61    63    45    47    54    84    35
  
[remainder of observations omitted]
Coefficients for equation 3.12, page 57.
proc reg data=p054;
  model y = x1 x2;
run;

The REG Procedure
Model: MODEL1
Dependent Variable: Y

                             Analysis of Variance

                                    Sum of           Mean
Source                   DF        Squares         Square    F Value    Pr > F
Model                     2     2935.10281     1467.55141      29.10    <.0001
Error                    27     1361.86385       50.43940
Corrected Total          29     4296.96667

Root MSE              7.10207    R-Square     0.6831
Dependent Mean       64.63333    Adj R-Sq     0.6596
Coeff Var            10.98825

                        Parameter Estimates

                     Parameter       Standard
Variable     DF       Estimate          Error    t Value    Pr > |t|
Intercept     1       15.32762        7.16023       2.14      0.0415
X1            1        0.78034        0.11939       6.54      <.0001
X2            1       -0.05016        0.12992      -0.39      0.7025
Coefficients for equation 3.13, page 57. Note: The output statement creates a new variable, eyx2, for the residual and places it along with the other variables into the SAS dataset p054b.
proc reg data=p054;
  model y = x1;
  output out=p054a residual=eyx1;
run;

The REG Procedure
Model: MODEL1
Dependent Variable: Y

                             Analysis of Variance

                                    Sum of           Mean
Source                   DF        Squares         Square    F Value    Pr > F
Model                     1     2927.58425     2927.58425      59.86    <.0001
Error                    28     1369.38241       48.90651
Corrected Total          29     4296.96667

Root MSE              6.99332    R-Square     0.6813
Dependent Mean       64.63333    Adj R-Sq     0.6699
Coeff Var            10.81999

                        Parameter Estimates

                     Parameter       Standard
Variable     DF       Estimate          Error    t Value    Pr > |t|
Intercept     1       14.37632        6.61999       2.17      0.0385
X1            1        0.75461        0.09753       7.74      <.000
Coefficients for equation 3.14, page 57.
proc reg data=p054a;
  model x2 = x1;
  output out=p054b residual=ex2x1;
run;

The REG Procedure
Model: MODEL1
Dependent Variable: X2

                             Analysis of Variance

                                    Sum of           Mean
Source                   DF        Squares         Square    F Value    Pr > F
Model                     1     1353.17314     1353.17314      12.68    0.0013
Error                    28     2988.29352      106.72477
Corrected Total          29     4341.46667

Root MSE             10.33077    R-Square     0.3117
Dependent Mean       53.13333    Adj R-Sq     0.2871
Coeff Var            19.44310

                        Parameter Estimates

                     Parameter       Standard
Variable     DF       Estimate          Error    t Value    Pr > |t|
Intercept     1       18.96540        9.77927       1.94      0.0626
X1            1        0.51303        0.14408       3.56      0.0013
Table 3.4, page 58.
proc print data=p054b;
  var eyx1 ex2x1;
run;

Obs        eyx1       ex2x1
  1    -10.6203    -15.1300
  2      0.2886     -0.7995
  3      4.4592     13.1224
  4     -1.2321     -6.2864
  5      7.6145     -2.9819
  6    -12.7887      1.8178
  7     -7.5039    -11.3385
  8     -0.3454     -7.4428
  9     -3.7043     10.9660
 10      6.3286     -5.2604
 11      9.9726      6.8439
 12      7.2093     -2.7473
 13      8.1502      6.2266
 14     -7.9329     21.4529
 15      4.2946     -4.4689
 16     -2.0505    -15.1383
 17     -4.4466      1.4269
 18      6.1122     15.2527
 19     -2.6443     -8.8776
 20     -7.1767     19.2787
 21      5.1139     -6.4867
 22      3.6797      1.7396
 23    -11.2220     -0.8255
 24     -2.0936      4.0524
 25      7.6405     -4.6691
 26     -6.1035      7.5311
 27      7.0559      0.5572
 28     -9.6002     -4.2082
 29      6.9045      8.4269
 30      4.6405    -22.0340
Coefficients for equation 3.15, page 58. Note: Values such as -4.6735E-17 are zero as represented by SAS in double precision.
proc reg data=p054b;
  model eyx1= ex2x1;
  run;
quit;

The REG Procedure
Model: MODEL1
Dependent Variable: eyx1 Residual

                             Analysis of Variance

                                    Sum of           Mean
Source                   DF        Squares         Square    F Value    Pr > F
Model                     1              0              0       0.00    1.0000
Error                    28     1361.86385       48.63799
Corrected Total          29     1361.86385

Root MSE              6.97409    R-Square     0.0000
Dependent Mean    -1.0244E-14    Adj R-Sq    -0.0357
Coeff Var         -6.80821E16

                               Parameter Estimates

                                  Parameter       Standard
Variable     Label        DF       Estimate          Error    t Value    Pr > |t|
Intercept    Intercept     1    -1.0244E-14        1.27329      -0.00      1.0000
ex2x1        Residual      1    -4.6735E-17        0.12758      -0.00      1.0000
Equation 3.25, page 62; Table 3.5, page 63; Table 3.7, page 67; and Equation 3.40, page 68. Note: The test statement is used to test specific hypotheses concerning regression parameters.
proc reg data=p054;
  model y = x1 x2 x3 x4 x5 x6;
  t1: test x2=0, x4=0, x5=0, x6=0;
run;

The REG Procedure
Model: MODEL1
Dependent Variable: Y

                             Analysis of Variance

                                    Sum of           Mean
Source                   DF        Squares         Square    F Value    Pr > F
Model                     6     3147.96634      524.66106      10.50    <.0001
Error                    23     1149.00032       49.95654
Corrected Total          29     4296.96667

Root MSE              7.06799    R-Square     0.7326
Dependent Mean       64.63333    Adj R-Sq     0.6628
Coeff Var            10.93552

                        Parameter Estimates

                     Parameter       Standard
Variable     DF       Estimate          Error    t Value    Pr > |t|
Intercept     1       10.78708       11.58926       0.93      0.3616
X1            1        0.61319        0.16098       3.81      0.0009
X2            1       -0.07305        0.13572      -0.54      0.5956
X3            1        0.32033        0.16852       1.90      0.0699
X4            1        0.08173        0.22148       0.37      0.7155
X5            1        0.03838        0.14700       0.26      0.7963
X6            1       -0.21706        0.17821      -1.22      0.2356

        Test T1 Results for Dependent Variable Y

                                Mean
Source             DF         Square    F Value    Pr > F
Numerator           4       26.41216       0.53    0.7158
Denominator        23       49.95654
Table 3.8, page 69.
proc reg data=p054;
  model y = x1 x3;
run;

The REG Procedure
Model: MODEL1
Dependent Variable: Y

                             Analysis of Variance

                                    Sum of           Mean
Source                   DF        Squares         Square    F Value    Pr > F
Model                     2     3042.31770     1521.15885      32.74    <.0001
Error                    27     1254.64897       46.46848
Corrected Total          29     4296.96667

Root MSE              6.81678    R-Square     0.7080
Dependent Mean       64.63333    Adj R-Sq     0.6864
Coeff Var            10.54685

                        Parameter Estimates

                     Parameter       Standard
Variable     DF       Estimate          Error    t Value    Pr > |t|
Intercept     1        9.87088        7.06122       1.40      0.1735
X1            1        0.64352        0.11848       5.43      <.0001
X3            1        0.21119        0.13440       1.57      0.1278
Equation following 3.48, page 71. 
Note 1: New variables need to be created in a data step. 
Note 2: The set statement is used to include all of the variables from the p054 dataset.
data p054c;
  set p054;
  w = x1 + x3;
run;

proc reg data=p054c;
  model y = w;
run;
quit;

The REG Procedure
Model: MODEL1
Dependent Variable: Y

                             Analysis of Variance

                                    Sum of           Mean
Source                   DF        Squares         Square    F Value    Pr > F

Model                     1     2872.37206     2872.37206      56.46    <.0001
Error                    28     1424.59461       50.87838
Corrected Total          29     4296.96667


Root MSE              7.13291    R-Square     0.6685
Dependent Mean       64.63333    Adj R-Sq     0.6566
Coeff Var            11.03596


                        Parameter Estimates

                     Parameter       Standard
Variable     DF       Estimate          Error    t Value    Pr > |t|

Intercept     1        9.98821        7.38841       1.35      0.1872
w             1        0.44439        0.05914       7.51      <.0001
F-ratio middle of page 71.
proc reg data=p054c;
  model y = x1 x3;
  t2: test x1 = x3;
run;
quit;
        Test t2 Results for Dependent Variable Y

                                Mean
Source             DF         Square    F Value    Pr > F

Numerator           1      169.94564       3.66    0.0665
Denominator        27       46.46848
F-ratio near the top of page 72.
proc reg data=p054c;
  model y = x1 x2 x3 x4 x5 x6;
  t3: test x1 = x3, x2=0, x4=0, x5=0, x6=0;
run;

[some output omitted]

The REG Procedure
Model: MODEL1

        Test T3 Results for Dependent Variable Y

                                Mean
Source             DF         Square    F Value    Pr > F
Numerator           5       55.11886       1.10    0.3857
Denominator        23       49.95654
Method 1: Equations at the bottom of page 72 and top of page 73. Note: The coefficient for x3, .306, needs to be computed by hand.
data p054d;
  set p054;
  yprime = y - x3;
  v = x1 - x3;
run;

proc reg data=p054d;
  model yprime = v;
run;

The REG Procedure
Model: MODEL1
Dependent Variable: yprime

                             Analysis of Variance

                                    Sum of           Mean
Source                   DF        Squares         Square    F Value    Pr > F

Model                     1     1794.31921     1794.31921      37.79    <.0001
Error                    28     1329.54746       47.48384
Corrected Total          29     3123.86667

Root MSE              6.89085    R-Square     0.5744
Dependent Mean        8.26667    Adj R-Sq     0.5592
Coeff Var            83.35708

                        Parameter Estimates

                     Parameter       Standard
Variable     DF       Estimate          Error    t Value    Pr > |t|
Intercept     1        1.16654        1.70788       0.68      0.5002
v             1        0.69382        0.11287       6.15      <.0001
Method 2: Equations at the bottom of page 72 and top of page 73. 
Note: The restrict command is used to compute constrained linear regressions. In this example, it computes the coefficients for both x1 and x3 in a single command.
proc reg data=p054;
  model y = x1 x3;
  restrict x1 + x3 = 1;
run;

The REG Procedure
Model: MODEL1
Dependent Variable: Y

NOTE: Restrictions have been applied to parameter estimates.

                             Analysis of Variance

                                    Sum of           Mean
Source                   DF        Squares         Square    F Value    Pr > F
Model                     1     2967.41921     2967.41921      62.49    <.0001
Error                    28     1329.54746       47.48384
Corrected Total          29     4296.96667

Root MSE              6.89085    R-Square     0.6906
Dependent Mean       64.63333    Adj R-Sq     0.6795
Coeff Var            10.66145

                        Parameter Estimates

                     Parameter       Standard
Variable     DF       Estimate          Error    t Value    Pr > |t|
Intercept     1        1.16654        1.70788       0.68     0.5002
X1            1        0.69382        0.11287       6.15     <.0001
X3            1        0.30618        0.11287       2.71     0.0113
RESTRICT     -1     -515.50829      410.46113      -1.26     0.2151*

* Probability computed using beta distribution.
F-ratio at the top of page 73.
proc reg data=p054;
  model y = x1 x3;
  t4: test x1 + x3 = 1;
run;

[some output omitted] 

The REG Procedure
Model: MODEL1

        Test T4 Results for Dependent Variable Y

                                Mean
Source             DF         Square    F Value    Pr > F
Numerator           1       74.89848       1.61    0.2151
Denominator        27       46.46848

How to cite this page

Report an error on this page

UCLA Researchers are invited to our Statistical Consulting Services
We recommend others to our list of Other Resources for Statistical Computing Help
These pages are Copyrighted (c) by UCLA Academic Technology Services


The content of this web site should not be construed as an endorsement of any particular web site, book, or software product by the University of California