IB close different from NYSE

P

pepper_john

Today (2012-01-13) NYSE close price for SPY is quite different from those recorded on IB and I hope someone can shed light on this. On NYSE the close price was 128.84, but on IB the close was 128.95 (the exchange is ARCA).

The 1-sec data on IB just before the market closed was

date time open high low close
20120113 15:59:50 128.86 128.87 128.84 128.85
20120113 15:59:51 128.85 128.86 128.84 128.85
20120113 15:59:52 128.85 128.86 128.85 128.86
20120113 15:59:53 128.86 128.88 128.85 128.87
20120113 15:59:54 128.88 128.88 128.85 128.87
20120113 15:59:55 128.87 128.89 128.87 128.89
20120113 15:59:56 128.88 128.89 128.88 128.89
20120113 15:59:57 128.89 128.89 128.87 128.88
20120113 15:59:58 128.88 128.9 128.88 128.89
20120113 15:59:59 128.89 128.96 128.89 128.95
20120113 16:00:00 128.95 129.05 128.95 129.03
20120113 16:00:01 129.02 129.15 129.02 129.15
20120113 16:00:02 129.14 129.14 129.07 129.1
20120113 16:00:03 129.08 129.08 128.9 129.04
20120113 16:00:04 129.08 129.09 129.03 129.05
20120113 16:00:05 129.05 129.07 129.03 129.04
20120113 16:00:06 129.05 129.15 128.84 129.12
20120113 16:00:07 129.12 129.12 129.04 129.06

Thanks.

john
 
See what price shows up on your statement. It will probably match the NYSE closing price. The TWS last price includes after hours trading. So no closing price is displayed until they reset their system overnight.
 
Thanks. I will check tomorrow for the close price. In any case, the difference between the IB price 128.95 at 16:00:00 seems to be too far away from NYSE close 128.84.
 
There are different interpretations of "Close".

The settlement value and the last trade are not necessarily equal.

IB usually reports the last trade in historical data.

The exchange uses the settlement value.

These may differ if a trade is later corrected, or cancelled, etc. Another example is of there was no trade right at the bell (this may happen in less active markets).

Of course IB will use the final settlement value in daily account reports.

Even in daily historical data on futures, I find that IB and ThinkOrSwim both report only the last trade and not the settlement value. Tradestation does report the true daily settlement value.

This is a real problem in futures markets where there there are no trades in certain options strikes etc. The IB data is unusable in this case because, lacking settlement data, it reports no value at all.
 
Are you asking for the composite closing price because of automatic option exercise, or because you had a market on close order in the SPY? For options, the composite close appears to be 128.84. If you entered a MOC order, your not entitled to the composite close, but the close on the exchange you routed your order to. They are not all the same.

BTW: The official composite close can be a trade printed up to 90 seconds after the close.
 
Quote from rmorse:

Are you asking for the composite closing price because of automatic option exercise, or because you had a market on close order in the SPY? For options, the composite close appears to be 128.84. If you entered a MOC order, your not entitled to the composite close, but the close on the exchange you routed your order to. They are not all the same.

BTW: The official composite close can be a trade printed up to 90 seconds after the close.
More...

Thanks. I looked at the IB's after-hours trades yesterday and they were consistently far above 128.84. So IB's AH data was probably not so accurate yesterday. Also yesterday yahoo and NYSE all reported the same close price of 128.84.

I wounder if the so-called "in-kind transaction", a feature unique to ETF, played a role in the close pricing of SPY. "In-kind transaction" is discussed here

http://www.nasdaq.com/indexshares/eqqqETFsInDetail.stm
 
Quote from comintel:

This is a real problem in futures markets where there there are no trades in certain options strikes etc. The IB data is unusable in this case because, lacking settlement data, it reports no value at all.
More...

The IB data represent what actually happened in the marketplace -- in some cases, nothing. Exchange settlements are in such cases are mark-to-model, not reports of actual transactions. You can't get blood from a turnip. If there's no volume, the exchange uses a model, i.e. a guess. And they're sometimes biased, because of exchange member internal politics.
 
Quote from pepper_john:

Thanks. I looked at the IB's after-hours trades yesterday and they were consistently far above 128.84. So IB's AH data was probably not so accurate yesterday. Also yesterday yahoo and NYSE all reported the same close price of 128.84.

I wounder if the so-called "in-kind transaction", a feature unique to ETF, played a role in the close pricing of SPY. "In-kind transaction" is discussed here

http://www.nasdaq.com/indexshares/eqqqETFsInDetail.stm
More...

The S&P 500 futures stay open until 3:15 (CT). After 3:00 these futures rallied almost $5, which would represent an almost .50 "after hours" rally in the SPY between 3 and 3:15
 
Quote from pepper_john:

Thanks. I looked at the IB's after-hours trades yesterday and they were consistently far above 128.84. So IB's AH data was probably not so accurate yesterday. Also yesterday yahoo and NYSE all reported the same close price of 128.84.

I wounder if the so-called "in-kind transaction", a feature unique to ETF, played a role in the close pricing of SPY. "In-kind transaction" is discussed here

http://www.nasdaq.com/indexshares/eqqqETFsInDetail.stm
More...


It's my understanding these transactions are not printed. They are submitted to the ETF administrator from a Broker Dealer. They are not transacted on an exchange.
 
Quote from Rodney King:

The IB data represent what actually happened in the marketplace -- in some cases, nothing. Exchange settlements are in such cases are mark-to-model, not reports of actual transactions. You can't get blood from a turnip. If there's no volume, the exchange uses a model, i.e. a guess. And they're sometimes biased, because of exchange member internal politics.
More...

I am aware of that but for daily historical data series in less-traded futures and especially futures options, settlement values are far superior.

Otherwise you have stretches of weeks or longer where particular futures months or option strikes are not traded at all or maybe once a day and there is no data or bad data in the historical data series or plot from IB.

This makes it absolutely useless to plot futures option values over a month or whatever - the plot is empty. This applies especially to out of the money futures options but also to some futures months.

Tradestation does it right and so do most of the larger quote vendors - they use the settlement value in daily data.
 
You must log in or register to reply here.
Back
Top