In your bit link, I trust you address that your first alternative - 'bottom-up' approach, uses the data to allow the analyst to intuitively formulate a model and thus hypothesis. And that your latter approach - 'top down' - depends on that intuitive thinking from prior data to conceive a model that is close enough to 'reality' for an hypothesis to handle the data.
I'm not sure that either extreme is effective, much less functional.
Jay
On Nov 29, 2013, at 8:32 PM, Vincent Granville <[log in to unmask]> wrote:
> There are two approaches to working with data: Data science, a bottom-up approach (from data to hypothesis); Statistics, a top-down approach (model-driven, from hypothesis to data). I disagree with this. I believe you can blend bottom-up and top-down approaches. As a data scientist, this is my philosophy.
>
> Read more at http://bit.ly/1bw7dzS
>
> You may leave the list at any time by sending the command
>
> SIGNOFF allstat
>
> to [log in to unmask], leaving the subject line blank.
Jay Warner
Working out of Racine, WI
On the road
[log in to unmask]
The A2Q Method – What do you want to improve, today?
You may leave the list at any time by sending the command
SIGNOFF allstat
to [log in to unmask], leaving the subject line blank.
|