JxFk(x)dx о 0, (28)
respectively, and to verify afterwards that the optimal weight functions Fk satisfy the stronger
conditions (6) and (7).
Without loss of generality we may represent the functions Fk by their Fourier flexible
form
∞∞
Fk(x) α αo,k + ∑ αj,kcos(2jπx) + ∑ βj,ksin(2jπx)
Then by some tedious but straightforward calculations it can be shown that:
Lemma 6. The conditions (27), (28), (8), (9), and (10) now read as:
Fk(χ)dχ α αok - 0

;(χ)dχ - -F-∑ β
2π J=1 J
ffFk(x )Fm (y )min(x, y ) dxdy
8π2

÷Σ
J=1
βJ,kβJ,
0 if k ≠ m ,
∞
F(x ) [Fm (y ) dydx = -r- Σ
4 4 4 П 7=1
β
J, kt^J, m
Vj-1
∞
Σ
j=1
β
J,m J,k
[Fk(x )Fm (x ) dx = ʌ
∞
V α..α.
X-7 j, k j,
Ij'1
ʌ
-∞
÷ βjkβ rnt
0 if k ≠ m .
Combining (1) and the results of Lemma 6, we have
16
More intriguing information
1. The name is absent2. Imputing Dairy Producers' Quota Discount Rate Using the Individual Export Milk Program in Quebec
3. The name is absent
4. Whatever happened to competition in space agency procurement? The case of NASA
5. Behavior-Based Early Language Development on a Humanoid Robot
6. The value-added of primary schools: what is it really measuring?
7. Structural Breakpoints in Volatility in International Markets
8. Transfer from primary school to secondary school
9. Sectoral Energy- and Labour-Productivity Convergence
10. Quality practices, priorities and performance: an international study