1NIST/ITL StRD
2Dataset Name:  Rat42             (Rat42.dat)
3
4File Format:   ASCII
5               Starting Values   (lines 41 to 43)
6               Certified Values  (lines 41 to 48)
7               Data              (lines 61 to 69)
8
9Procedure:     Nonlinear Least Squares Regression
10
11Description:   This model and data are an example of fitting
12               sigmoidal growth curves taken from Ratkowsky (1983).
13               The response variable is pasture yield, and the
14               predictor variable is growing time.
15
16
17Reference:     Ratkowsky, D.A. (1983).
18               Nonlinear Regression Modeling.
19               New York, NY:  Marcel Dekker, pp. 61 and 88.
20
21
22
23
24
25Data:          1 Response  (y = pasture yield)
26               1 Predictor (x = growing time)
27               9 Observations
28               Higher Level of Difficulty
29               Observed Data
30
31Model:         Exponential Class
32               3 Parameters (b1 to b3)
33
34               y = b1 / (1+exp[b2-b3*x])  +  e
35
36
37
38          Starting Values                  Certified Values
39
40        Start 1     Start 2           Parameter     Standard Deviation
41  b1 =   100         75            7.2462237576E+01  1.7340283401E+00
42  b2 =     1          2.5          2.6180768402E+00  8.8295217536E-02
43  b3 =     0.1        0.07         6.7359200066E-02  3.4465663377E-03
44
45Residual Sum of Squares:                    8.0565229338E+00
46Residual Standard Deviation:                1.1587725499E+00
47Degrees of Freedom:                                6
48Number of Observations:                            9
49
50
51
52
53
54
55
56
57
58
59
60Data:   y              x
61       8.930E0        9.000E0
62      10.800E0       14.000E0
63      18.590E0       21.000E0
64      22.330E0       28.000E0
65      39.350E0       42.000E0
66      56.110E0       57.000E0
67      61.730E0       63.000E0
68      64.620E0       70.000E0
69      67.080E0       79.000E0
70