Subject:
|
Mechanics of proxy bidding?
|
Newsgroups:
|
lugnet.market.theory
|
Date:
|
Thu, 28 Oct 1999 18:11:36 GMT
|
Viewed:
|
364 times
|
| |
|
|
I'm trying to figure out how to do proxy bidding (with multiple instances of
a lot). My current assumptions are:
- Each instance of a lot stores the following information: the bidder, the
max. proxy bid placed by that bidder, and the timestamp of the bid (and
of course the current price of the instance)
- Instances are stored/listed in order, where the primary order is
determined by asking price and secondary order is determined by timestamp
just like AucZILLA, and my software, currently do it). "Heavier" (pricier,
older) instances sink to the "bottom" of the list.
- Bid negotiation rules are as follows:
- Higher price wins
- If prices are the same, older timestamp wins (so older proxy bids
win out against newer bids of the same value)
- As soon as a bump occurs, the instances are re-ordered if necessary to
maintain the ordering described above.
- If a bid is bumped, and there was a higher proxy bid on that instance,
the proxy bid is automatically re-submitted to the system, just as if the
bidder had submitted it (except that the older timestamp stays intact).
This does not always achieve the desired outcome. Consider this example
with two instances of a lot ("foo"); the second price column denotes
current, not asking, price.
Starting state (A's bid is older than B's):
foo 1.00} .50 B
foo .80} .50 A
C bids 0.60, bumping B, and the instances reorder, based on price:
foo .80} .50 A
foo .60} .60 C
B rebids 1.00, bumping A, and the instances reorder, based on timestamp:
foo .60} .60 C
foo 1.00} .60 B
A rebids .80, bumping C, and the instances reorder, based on price:
foo 1.00} .60 B
foo .80} .70 A
A's bid was older than B's, so A shouldn't end up paying more than B. What
am I missing here?
I'm hoping that Todd in particular will offer some hints about this, but
other insights are welcome as well.
Steve
--
Barb & Steve Demlow | demlow@visi.com | www.visi.com/~demlow/
|
|
Message has 1 Reply:
 | | Re: Mechanics of proxy bidding?
|
| (...) One thing you need to decide is if a proxy need only meet the bid which causes it to be re-submitted, or to beat it by a bid increment if possible. It looks like you're planning on the 2nd. The way I would be inclined to handle the situation (...) (25 years ago, 28-Oct-99, to lugnet.market.theory)
|
4 Messages in This Thread:
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
