Multinomial Regression
This example used Minitab 12 saved data set, tvhrs.mtw. Inside Minitab, go into the Minitab folder and click on Student12 à tvhrs.mtw . For this data, we will try to predict the amount of hours spent watching mtv based on the age of the participant, the number of hours spent watching tv, and the number of hours spent watching the news.
Getting Slope, Y Intercept and Model Utility Test

Regression Analysis: HrsMTV versus
Age, HrsTV, HrsNews
The
regression equation is
HrsMTV =
1.15 - 0.0439 Age + 0.125 HrsTV - 0.101 HrsNews
Predictor Coef SE Coef T P
Constant 1.1547 0.4258 2.71 0.008
Age
-0.043896 0.009849 -4.46 0.000
HrsTV
0.12492 0.02101 5.94 0.000
HrsNews -0.10064 0.05630 -1.79 0.076
S =
2.25002 R-Sq = 30.8% R-Sq(adj) = 29.0%
Analysis of
Variance
Source
DF
SS
MS
F P
Regression
3 261.329 87.110 17.21 0.000
Residual
Error 116 587.263 5.063
Total 119 848.592
Source DF Seq SS
Age 1 81.761
HrsTV 1 163.387
HrsNews 1 16.180

3.
Then, select Four in One under Residual Plots and put in all of
your x’s in the Residual versus variable box.
Note: This will generate a lot of graphs. If you want
to close all the graphs, go to Windows- à
Close All Windows.
First, label a column
for the squared term. Click on Calc, Calculator. The following screen will
appear. Put the name of the new column in the first box called Store result
in variable.

In
the expression box, enter the expression that you want it to calculate. For
instance, HrsTV squared would be HrsTV*HrsTV.
Click o.k. Your new
variable is in the column.
Prediction Interval for new y and Confidence interval for average y at a combination of x values.
Follow the steps above but before you hit o.k. Go
to options. The following screen should appear.

In the box for Prediction Intervals for new
observations, enter your new x's. For instance, if you want age to be 21,
hrstv to be 7 and hrsnews to be 1, enter 21 7 1 into the Prediction intervals for new observations
box. Make sure that you enter them in the same order as they appear in the
predictors box on the previous screen.
If you want to change your confidence, enter that
as well in the Confidence level box below.
Click O.K.
Additional lines will be added to your output. It
should look similar to the following output below.
This will give you both the confidence interval
and the prediction interval.
Predicted
Values for New Observations
New
Obs Fit SE Fit 95% CI 95% PI
1
1.007 0.301 (0.410, 1.604) (-3.490, 5.503)
Values of
Predictors for New Observations
New
Obs Age HrsTV HrsNews
1
21.0 7.00 1.00
Prediction Interval for new y and Confidence interval for average y for the entire data set.
Follow the steps above but before you hit o.k. Go
to options. The following screen should appear.

In the box for Prediction Intervals for new
observations, double click on the variables listed to the left that appear in
the model. Make sure that you list them in the same order as you listed them in
the Prediction intervals for new observations window.
If you want to change your confidence, enter that
as well in the Confidence level box below.
Click O.K.
Additional lines will be added to your output. It should
look similar to the following output below.
This will give you both the confidence interval
and the prediction interval for every item in the original data set. .
Predicted
Values for New Observations
New Obs Fit SE Fit 95% CI
95% PI
1 2.228 0.252 (
1.729, 2.727) (-2.256, 6.712)
2 0.982 0.329 (
0.331, 1.634) (-3.521, 5.486)
3 1.600 0.413 (
0.782, 2.418) (-2.931, 6.131)
4 1.281 0.264 (
0.758, 1.803) (-3.206, 5.768)
5 1.185 0.263 (
0.665, 1.705) (-3.302, 5.672)
6 0.860 0.256 (
0.353, 1.367) (-3.625, 5.345)
7
1.487 0.251 ( 0.989, 1.985)
(-2.997, 5.971)
8 1.080 0.267 (
0.550, 1.609) (-3.408, 5.567)
9 1.036 0.264 (
0.512, 1.559) (-3.451, 5.523)
10 0.830 0.290 (
0.256, 1.404) (-3.664, 5.323)
11 0.382 0.360
(-0.331, 1.095) (-4.131, 4.895)
12 1.962 0.273 (
1.422, 2.503) (-2.527, 6.451)
13 0.805 0.292 (
0.228, 1.383) (-3.688, 5.299)
14 0.955 0.278 (
0.404, 1.505) (-3.536, 5.445)
15 0.257 0.374
(-0.483, 0.998) (-4.260, 4.775)
16 0.979 0.286 (
0.412, 1.545) (-3.513, 5.471)
17 1.338 0.279 (
0.784, 1.891) (-3.153, 5.828)
18 2.228 0.252 (
1.729, 2.727) (-2.256, 6.712)
19 0.169 0.375
(-0.574, 0.913) (-4.349, 4.687)
20 0.781 0.303 (
0.182, 1.380) (-3.715, 5.278)
21 1.435 0.248 (
0.943, 1.926) (-3.049, 5.918)
22 1.381 0.280 (
0.828, 1.935) (-3.109, 5.872)
23 0.257 0.374
(-0.483, 0.998) (-4.260, 4.775)
24 1.180 0.259 (
0.666, 1.694) (-3.306, 5.666)
25 1.036 0.264 (
0.512, 1.559) (-3.451, 5.523)
26 1.560 0.243 (
1.077, 2.042) (-2.923, 6.042)
27 1.330 0.458 (
0.423, 2.237) (-3.217, 5.878)
28
1.805 0.236 ( 1.338, 2.272)
(-2.676, 6.286)
29 1.285 0.246 (
0.798, 1.773) (-3.198, 5.769)
30 0.882 0.311 (
0.266, 1.498) (-3.617, 5.381)
31 2.071 0.300 (
1.476, 2.666) (-2.425, 6.567)
32 0.720 0.334 (
0.059, 1.381) (-3.785, 5.225)
33 1.780 0.249 (
1.288, 2.273) (-2.703, 6.264)
34 1.136 0.257 (
0.626, 1.646) (-3.349, 5.622)
35 1.535 0.235 (
1.071, 2.000) (-2.945, 6.016)
36 1.681 0.404 (
0.881, 2.481) (-2.847, 6.209)
37 2.507 0.338 (
1.837, 3.176) (-2.000, 7.013)
38 2.897 0.301 (
2.300, 3.493) (-1.600, 7.393)
39 1.603 0.248 (
1.112, 2.095) (-2.880, 6.087)
40 1.031 0.281 (
0.475, 1.587) (-3.460, 5.522)
41 2.076 0.465 (
1.156, 2.996) (-2.475, 6.627)
42 3.079 0.333 (
2.419, 3.739) (-1.426, 7.584)
43 1.733 0.293 (
1.152, 2.314) (-2.762, 6.227)
44 3.082 0.355 (
2.379, 3.786) (-1.429, 7.594)
45 2.232 0.289 (
1.660, 2.805) (-2.261, 6.725)
46 2.531 0.280 (
1.975, 3.086) (-1.960, 7.022)
47 2.579 0.303 (
1.979, 3.179) (-1.917, 7.076)
48 1.330 0.324 (
0.688, 1.972) (-3.172, 5.833)
49 1.882
0.277 ( 1.334, 2.430) (-2.608, 6.372)
50 2.998 0.331 (
2.342, 3.654) (-1.507, 7.502)
51 4.831 0.561 (
3.720, 5.943) ( 0.238, 9.424)
52 2.188 0.286 (
1.622, 2.755) (-2.304, 6.681)
53 4.130 0.440 (
3.259, 5.002) (-0.411, 8.671)
54 3.579 0.380 (
2.825, 4.332) (-0.941, 8.098)
55 2.083 0.310 (
1.470, 2.696) (-2.415, 6.582)
56 3.981 0.432 (
3.126, 4.836) (-0.557, 8.519)
57 2.025 0.938 (
0.168, 3.882) (-2.803, 6.853)XX
58 0.939 0.339 (
0.268, 1.611) (-3.568, 5.446)
59 0.939 0.339 (
0.268, 1.611) (-3.568, 5.446)
60 0.939 0.339 (
0.268, 1.611) (-3.568, 5.446)
61 2.188 0.286 (
1.622, 2.755) (-2.304, 6.681)
62 2.232 0.289 (
1.660, 2.805) (-2.261, 6.725)
63 1.531 0.288 (
0.961, 2.102) (-2.961, 6.024)
64 4.409 0.482 (
3.455, 5.364) (-0.148, 8.967)
65 6.573 0.864 (
4.861, 8.285) ( 1.799, 11.347)X
66 3.793 0.400 (
3.000, 4.585) (-0.734, 8.319)
67 5.787 0.705 (
4.390, 7.183) ( 1.117, 10.457)
68 1.689 0.290 (
1.113, 2.264) (-2.805, 6.182)
69 1.588 0.281 (
1.031, 2.145) (-2.903, 6.079)
70 4.755 0.526 (
3.713, 5.797) ( 0.178, 9.332)
71 1.113 0.311 (
0.497, 1.728) (-3.386, 5.611)
72 3.082 0.355 (
2.379, 3.786) (-1.429, 7.594)
73 1.338 0.296 (
0.752, 1.924) (-3.157, 5.833)
74 1.632 0.285 (
1.067, 2.197) (-2.860, 6.124)
75 1.588 0.281 (
1.031, 2.145) (-2.903, 6.079)
76 1.435 0.332 (
0.778, 2.092) (-3.069, 5.940)
77 1.628 0.361 (
0.914, 2.342) (-2.885, 6.142)
78 2.949 0.308 (
2.338, 3.560) (-1.549, 7.447)
79 3.207 0.366 (
2.483, 3.932) (-1.308, 7.722)
80 1.334 0.332 (
0.676, 1.991) (-3.171, 5.838)
81 2.379 0.520 (
1.350, 3.408) (-2.195, 6.953)
82 -0.631 0.408
(-1.438, 0.177) (-5.160, 3.898)
83 0.330 0.352
(-0.368, 1.028) (-4.181, 4.841)
84 1.592 0.419 (
0.763, 2.422) (-2.941, 6.125)
85 -0.741 0.497
(-1.725, 0.243) (-5.305, 3.823)
86 1.610 0.450 (
0.719, 2.501) (-2.935, 6.154)
87
1.634 0.403 ( 0.836, 2.432)
(-2.894, 6.161)
88 -0.067 0.597
(-1.250, 1.116) (-4.678, 4.544)
89 0.192 0.302
(-0.406, 0.790) (-4.304, 4.688)
90 -0.027 0.339
(-0.699, 0.644) (-4.534, 4.479)
91 -0.935 0.414
(-1.756, -0.115)
(-5.466, 3.596)
92 -1.044 0.423
(-1.881, -0.206)
(-5.578, 3.491)
93 1.770 0.531 (
0.718, 2.821) (-2.809, 6.349)
94 1.690 0.737 (
0.230, 3.151) (-2.999, 6.380)X
95 1.580 0.541 (
0.509, 2.652) (-3.003, 6.164)
96 1.282 0.455 (
0.382, 2.183) (-3.264, 5.829)
97 0.028 0.563
(-1.087, 1.143) (-4.566, 4.622)
98 4.561 0.845 (
2.888, 6.234) (-0.199, 9.321)X
99 -0.257 0.373
(-0.997, 0.482) (-4.775, 4.260)
100 -0.382 0.379
(-1.133, 0.368) (-4.901, 4.137)
101 -0.709 0.488
(-1.675, 0.257) (-5.269, 3.851)
102 -0.544 0.383
(-1.302, 0.214) (-5.065, 3.976)
103 0.020 0.374
(-0.720, 0.761) (-4.497, 4.538)
104 -0.867 0.438
(-1.734, 0.001) (-5.407, 3.673)
105 0.083 0.468
(-0.844, 1.010) (-4.469, 4.635)
106 -0.665 0.481
(-1.618, 0.287) (-5.222, 3.892)
107 -0.924 0.394
(-1.705, -0.142)
(-5.448, 3.601)
108 -0.764 0.416
(-1.588, 0.060) (-5.296, 3.768)
109 0.440 0.296
(-0.146, 1.026) (-4.055, 4.935)
110 -0.495 0.629
(-1.742, 0.751) (-5.123, 4.132)
111 1.404 0.424 (
0.564, 2.244) (-3.131, 5.939)
112 1.009 0.721
(-0.419, 2.437) (-3.671, 5.689)X
113 -1.381 0.775
(-2.917, 0.154) (-6.095, 3.332)X
114 -0.546 0.429
(-1.397, 0.304) (-5.083, 3.991)
115 3.549 0.638 (
2.285, 4.813) (-1.083, 8.181)
116
-2.346 0.738 (-3.809, -0.884) (-7.036, 2.344)X
117 0.833 0.389 (
0.063, 1.604) (-3.689, 5.356)
118 -0.036 0.354
(-0.738, 0.666) (-4.548, 4.475)
119 -0.467 0.488
(-1.434, 0.499) (-5.027, 4.093)
120 0.297 0.360 (-0.417, 1.010)
(-4.217, 4.810)
X denotes a
point that is an outlier in the predictors.
XX denotes a
point that is an extreme outlier in the predictors.
Values of
Predictors for New Observations
New Obs Age HrsTV HrsNews
1 21.0 20.0 5.0
2 21.0 6.0 0.0
3 21.0 19.0 10.0
4 21.0 10.0 2.0
5 22.0 12.0 5.0
6 26.0 10.0 4.0
7 22.0 12.0 2.0
8 21.0 10.0 4.0
9 22.0 10.0 4.0
10 21.0 8.0 4.0
11 21.0 2.0 1.0
12 22.0 15.0 1.0
13 21.0 7.0 3.0
14 21.0 9.0 4.0
15 21.0 1.0 1.0
16 21.0 10.0 5.0
17 22.0 10.0 1.0
18
21.0 20.0 5.0
19 23.0 1.0 1.0
20 21.0 6.0 2.0
21 22.0 14.0 5.0
22 21.0 10.0 1.0
23 21.0 1.0 1.0
24 21.0 10.0 3.0
25 22.0 10.0 4.0
26 22.0 15.0 5.0
27 22.0 18.0 11.0
28 21.0 15.0 3.0
29 22.0 12.0 4.0
30 21.0 6.0 1.0
31 20.0 20.0 7.0
32 19.0 4.0 1.0
33 21.0 14.0 2.0
34 22.0 10.0 3.0
35 22.0 14.0 4.0
36 22.0 20.0 10.0
37 22.0 25.0 8.0
38 20.0 25.0 5.0
39 21.0 15.0 5.0
40 21.0 8.0 2.0
41 13.0 20.0 10.0
42 13.0 24.0 5.0
43 13.0 10.0 1.0
44 13.0 20.0 0.0
45 13.0 14.0 1.0
46 13.0 18.0 3.0
47 13.0 20.0 5.0