Implementation in Julia

TimeSeries is a registered package. So like other packages, we can add it to your Julia packages:

Pkg.update() 
Pkg.add("TimeSeries")

The TimeArray time series type

immutable TimeArray{T, N, D<:TimeType, A<:AbstractArray} <: AbstractTimeSeries

 timestamp::Vector{D}
 values::A
 colnames::Vector{UTF8String}
 meta::Any

 function TimeArray(timestamp::Vector{D},
 values::AbstractArray{T,N},
 colnames::Vector{UTF8String},
 meta::Any)
 nrow, ncol = size(values, 1), size(values, 2)
 nrow != size(timestamp, 1) ? error("values must match length of
 timestamp"):
 ncol != size(colnames,1) ? error("column names must match width of
 array"):
 timestamp != unique(timestamp) ? error("there are duplicate dates"):
 ~(flipdim(timestamp, 1) == sort(timestamp) || timestamp ==
 sort(timestamp)) ? error("dates are mangled"):
 flipdim(timestamp, 1) == sort(timestamp) ? 
 new(flipdim(timestamp, 1), flipdim(values, 1), colnames, meta...