> From: Johann Hibschman
> Sent: Wednesday, 23 June 2010 02:39
>
> My basic goal is to read a CSV file into some kind of named associative
> structure, then do analysis on the columns of that data.
>
> Imagining that I have a csv file with time, price, and volume columns,
> in R, I can do:
>
> data <- read.csv("file.csv") # no assumptions on cols
> c(mean(data$price), sd(data$price))
>
> What would be the equivalent in J?
J probably isn't quite as set up for associative structures as R or q
nevertheless, assuming the first row contains names that can be interpreted as
valid J names:
require 'tables/csv stats/base'
data=: readcsv 'file.csv'
({.data) =: |: makenum }. Data
(mean , stddev) price
Note that 'tables/csv' and 'stats/base' are addons (like R packages) that you
download using Package Manager.
> Hopefully this is the right forum to ask this; if not, please point me
> in the right direction.
Probably better in the programming forum.
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
