Hello Julia – philip’s curve

Not sure how I landed up browsing through Julia for Data Science and Julia Data Science this Saturday evening. Julia is a fascinating language, but I think the learning curve is a bit much for me. Tried making a simple line chart and it’s … pretty hard, although the syntax is quite interesting …

begin
    using CSV, DataFrames
    using Downloads: download
    URL_DATA = "https://storage.dosm.gov.my/cpi/cpi_2d.csv"
    download(URL_DATA, "cpi_2d.csv")
    df = CSV.read("cpi_2d.csv", DataFrame)
end

begin
    using DataFramesMeta
    using AlgebraOfGraphics
    using CairoMakie
    using Dates: Year
    
    @rsubset! df :division == "overall"

end

dat = @chain df begin
    @select :date :lagged_index = :index
    @rtransform :date = :date + Year(1)
    rightjoin(df; on = :date)
    @rtransform :yoy = :index / :lagged_index - 1   
    dropmissing(:yoy)
end

begin
    set_aog_theme!()
    plt =  data(dat) * mapping(:date, :yoy) 
    layers = visual(Scatter) + visual(Lines)
    draw(
        plt * layers,
        axis = (
        ytickformat = "{:.0%}", 
        title = "Year-on-year change, consumer price index, Malaysia",
        xlabel = "", 
        ylabel = "")
    )
end