milktrader
milktrader
Why can't you just do ``` julia function add_vertex!{V1, V2}(g::GenericGraph{V1}, v::V2) ``` and ensure that V1 and V2 are valid types defined somewhere in the module?
Hmm, this question might be better answered on the julia-users mailing list. https://groups.google.com/forum/#!forum/julia-users I would try that forum.
Progress not so much, but thanks for re-raising this issue. It's not a big project really, just copy and paste from TimeSeries, MarketData, MarketTechnicals ...
Possibly ping @carljv about model hierarchy. Heads up on TimeSeries. It no longer supports DataFrames/DataArrays but instead has implemented the TimeArray type. ``` julia immutable TimeArray{T,N}
It does not yet deal with missing values. That missing piece is being explored in the DataArrays branch. The creation of a TimeArray is straightforward. ``` julia julia> mydates =...
If you don't mind cloning the `MarketData` package you get some `const` objects to play with. ``` julia julia> Pkg.clone("git://github.com/JuliaQuant/MarketData.jl.git") julia> using MarketData julia> OHLC 16103x6 TimeArray{Float64,2} 1950-01-03 to 2013-12-31...
Dropping down to just an array is simple ``` julia julia> OHLC.values 16103x6 Array{Float64,2}: 16.66 16.66 16.66 16.66 1.26e6 16.66 16.85 16.85 16.85 16.85 1.89e6 16.85 16.93 16.93 16.93 16.93...
@AndreyKolev I've looked into the best way to merge two repositories like these and it's actually not a simple matter. Given that the original TimeModels doesn't have a deep history...
@scidom I've been creating dummy timestamp data doing this. `[date(1,1,1):years(1):date(variable_that_has_length,1,1)]` So if you want 100 observations with dummy timestamps, you'd do: `[date(1,1,1):years(1):date(100,1,1)]`