regress
Perform multiple linear regression using the model y = X * beta + e.
Syntax
[b,bi,r,ri,stats]=regress(y,X)
[b,bi,r,ri,stats]=regress(y,X,alpha)
Inputs
- y
- The response values.
- X
- The regressor variable values. The first column is a vector of 1 values.
- alpha
- The level of significance (default: 0.05).
Outputs
- b
- The regression coefficients (beta estimates).
- bci
- The regression coefficient confidence intervals.
- r
- The residuals.
- rci
- The residuals confidence intervals.
- stats
- Regression statistics. The following values are returned:
Example
y = [6.59; 7.89; 8.49; 3.5; 6.7; 6.9; 4.99; 5.09; 8.09]
X = [1.0, -0.69, 2.0; 1.0, -0.69, 3.0; 1.0, -0.69, 4.0;
1.0, 0.0, 2.0; 1.0, 0.0, 3.0; 1.0, 0.0, 4.0;
1.0, 0.41, 2.0; 1.0, 0.41, 3.0; 1.0, 0.41, 4.0]
[b,bi,r,ri,stats]=regress(y,X)
b = [Matrix] 3 x 1
2.12192
-1.59846
1.40000
bi = [Matrix] 3 x 2
-0.82651 5.07035
-3.30187 0.10494
0.45306 2.34694
r = [Matrix] 9 x 1
0.56514
0.46514
-0.33486
-1.42192
0.37808
-0.82192
0.72345
-0.57655
1.02345
ri = [Matrix] 9 x 2
-1.17863 2.30891
-1.59401 2.52430
-2.14839 1.47867
-2.90897 0.06513
-1.97507 2.73122
-2.77729 1.13345
-1.05248 2.49938
-2.68810 1.53499
-0.56601 2.61290
stats = [Matrix] 1 x 4
0.75369 9.17986 0.01494 0.89858