LQ name conflict

[arnavs]

Now that Julia has an LQ type (in the LinearAlgebra stdlib), there's a name conflict with the type exported by the package.

One way to handle this would be to change the following to ... struct LQProblem with the appropriate constructor.

mutable struct LQ
    Q::ScalarOrArray
    R::ScalarOrArray
    A::ScalarOrArray
    B::ScalarOrArray
    C::ScalarOrArray
    N::ScalarOrArray
    bet::Real
    capT::Union{Int, Nothing} # terminal period
    rf::ScalarOrArray
    P::ScalarOrArray
    d::Real
    F::ScalarOrArray # policy rule
end

https://github.com/QuantEcon/lecture-source-jl/issues/436

[cc7768]

That's annoying...

What do people think about LQProgram rather than LQProblem?

[Nosferican]

From the docstring,

Provides a type called LQ for solving linear quadratic control problems.

Ideally we would move toward more descriptive names rather than 3-letter structs (definitely not two), but until then I think LQC should be fine. Other exported structs with accronyms are: ARMA, LAE, LSS, and RBLQ (probably good to adjust this one as well based on the change to LQ).

[arnavs]

In the short run I think LQC or LQP sounds good. Probably best to keep style consistent with the other types, etc.