√99以上 Y wAX^C 50ã Ê· 322457
In fact, provided the condition above holds (ie, there exists a feasible x with ¯cTx ≥ α) we can solve the problem (1) via convex optimization We make the change of variables y = x ¯cTx−α, s = 1 ¯cTx−α, so x = y/s This yields the problem minimize q yTRy subject toT = E(X tX 0) = a2E(cos(Θ)cos(λtΘ)) = a2E 1 2 {cos(λt2Θ)cos(λt)} = a2 2 1 2π Z 2π 0 cos(λt2θ)cos(λt)dθ = a2 2 cos(λt) So ρ t = cos(λt) The spectrum is F where γ t = R π −π eitω dF(ω) Try the discrete distribution for F, F(λ) = F(−λ) = c, a constant, F(ω) = 0 otherwise Then γ t = eitλce−itλc = ccos(tλ3 Is it possible for two independent random variables X and Y (not necessarily with the same distributions) to satisfy PX Projecteuclid Org "Y wAX^C 50'ã Ê'·