In: Statistics and Probability
Problem 4. The data {(1, 10),(2, 5.49),(3, 0.89),(4, −0.14),(5, −1.07),(6, 0.84)} comes from a model F(x) = (r/ x )+ sx. Use least squares to estimate the parameters r, s.
Given
Data Points: {(1, 10), (2, 5.49), (3, 0.89), (4, -0.14), (5, -1.07), (6, 0.84)}
In tabulated form:
i | 1 | 2 | 3 | 4 | 5 | 6 |
1 | 2 | 3 | 4 | 5 | 6 | |
10 | 5.49 | 0.89 | -0.14 | -1.07 | 0.84 |
Model:
Solution
Let the fitted values be
Now consider the residues,
Then, using least squares method, our aim is to minimize
To get the least square estimate, i.e., the least value of Q, we should ensure that:
And
Now,
And,
Solving, equations (1) and (2), we get:
At these values of r and s, the partial second order derivatives should be positive to ensure that Q is in fact a minimum value.
Computing the three partial second order derivatives:
As required, all these are positive.
Hence, the values of r and s are: