On seeing the attachments to Bland and Bibby's messages, I assumed the
Buses of York were proposed as a problem more insoluble than the Bridges
of Königsberg - which I'm sure they are. In fact, it's a nice
demonstration of the ubiquity of Missing Data in real-life statistics.
The viri ebor would do well to assure the divine emperor that they are
not slacking in his month, lest they, like the buses, are decimated.
Correlate four observations? Just don't. It's what I call the "informal
Bayesian" approach: if you had a strong theory before collecting the
data that they would be correlated, you've learned nothing; if they were
arbitrary variables you hadn't anticipated being related, you shouldn't
believe it.
And don't send out a press release!
Allan
On 25/07/2018 00:00, allstat automatic digest system wrote:
> *York Bus Forum: *
> Please note the following meetings, ...
> *17 July: *Monthly meeting
> *18 September: *Open Public Meeting to discuss our discussion paper "Better than a Bus Station"
You may leave the list at any time by sending the command
SIGNOFF allstat
to [log in to unmask], leaving the subject line blank.
|