Approximation algorithms

optimalportfolio_fptas(mkt, ε; verbose=false) -> X, v

Use the fully polynomial-time approximation scheme to produce a 1-ε-optimal portfolio X, with valuation v, for the VariedCostsMarket defined by mkt.

julia> mkt = VariedCostsMarket([0.2, 0.5, 0.1, 0.6, 0.1], [1, 4, 9, 1, 8], [2, 4, 2, 5, 1], 8);

julia> optimalportfolio_fptas(mkt, 0.2)
([3, 5, 2], 3.24)