Trade data w/o Bid/Ask quotes is useless/worthless

Q

Quanto

BEWARE:

If your data has only the trade data, but no Bid/Ask top-of-the-book quotes,
then your data cannot be used for making accurate trade decisions (be it for real trading or for backtesting).

Ie. the dataset should have also these data fields:
Bid, Ask, BidSize, AskSize, ...

I saw some (historic) options data sets by commercial vendors who do not have Bid/Ask quotes.
Such data is IMO useless and worthless... :)
...because of the Bid/Ask spreads... LastPrice (ie. the last trade) is not sufficient as it can be very old... One rather needs also the quotes.
 
Last edited:
Quanto said:
BEWARE:

If your data has only the trade data, but no Bid/Ask top-of-the-book quotes,
then your data cannot be used for making accurate trade decisions (be it for real trading or for backtesting).

Ie. the dataset should have also these data fields:
Bid, Ask, BidSize, AskSize, ...

I saw some (historic) options data sets by commercial vendors who do not have Bid/Ask quotes.
Such data is IMO useless and worthless... :)
...because of the Bid/Ask spreads... LastPrice (ie. the last trade) is not sufficient as it can be very old... One rather needs also the quotes.
More...


What is your intention, looking at managing tens of million dollar portfolios? or to work with some large firm?
 
Quanto said:
Aeh, what you mean? :) You sure replying in the right thread?
More...

May have read between the lines too much here, if so that's on me and I apologize. Via ignorance didn't realize Quanto, meant something completely different than Quant.

People seem to be so fascinated and interested in professional quants and they are near useless and over complicate things in general for retail trading. A quant would be useful if you're looking to make returns on large sums of money, no doubt. It's mainly for either finding short term niche to take advantage of in the market or to help manage larger amount of funds.

As far as actual trading, most quants or similar background suck it and as far as producing a long term edge non-erodible edge not very helpful.
 
Quanto said:
@Concinnity, you're getting off-topic... :)
More...


You can ask me to stay off your thread and I will. Otherwise I have to disagree. I would at least hope that more people are here to be profitable, make mutual beneficial relationships or to help others be profitable. So factual information about retail traders or related things(which is what majority of people here are) is never a waste of time. In fact information structured around that, might be the least off topic of anything.
 
Concinnity said:
You can ask me to stay off your thread and I will. Otherwise I have to disagree. I would at least hope that more people are here to be profitable, make mutual beneficial relationships or to help others be profitable. So factual information about retail traders or related things(which is what majority of people here are) is never a waste of time. In fact information structured around that, might be the least off topic of anything.
More...
Good, but then you should do it in an appropriate thread b/c this one here is about market data only.
Maybe you should open your own discussion thread in an appropriate board; I think this "data sets and feeds" section is not the right one for your discussion topic. Thx.
 
Last edited:
Quanto said:
BEWARE:

If your data has only the trade data, but no Bid/Ask top-of-the-book quotes,
then your data cannot be used for making accurate trade decisions (be it for real trading or for backtesting).

Ie. the dataset should have also these data fields:
Bid, Ask, BidSize, AskSize, ...

I saw some (historic) options data sets by commercial vendors who do not have Bid/Ask quotes.
Such data is IMO useless and worthless... :)
...because of the Bid/Ask spreads... LastPrice (ie. the last trade) is not sufficient as it can be very old... One rather needs also the quotes.
More...



The staleness of the LAST price can be seen easily in some of the
granular tick-increment micro futures products like MYM, M2K, and others.
Of important note, staleness of LAST price is not limited to only futures products, micro or not.

Further, I am not even suggesting those products are illiquid. I am saying that
due to the tick-increment granularity, liquidity at each tick-incrment may
be thin, or using a less derogatory word, differing, when compared to their non-micro counterparts.

As for your premise about worth/usefulness of B/A and B/A-size, I disagree.
DOM shenanigans is a real thing. Nonetheless, for some strategies, i do agree there may be worth.
I sometimes refer to SIZE data from T&S. That's a hindsight real-time data fact! LOL.
 
Also the folowing is important for trade decisions as well for backtestings:

The quote data MUST be from BEFORE market close!
B/c at and after market close the DAY orders are flushed (deleted) from the orderbook, then only GTC orders remain, but which can be very old, therefore with an unrealistic Bid/Ask... Ie, such unrealistic Bid/Ask is not usable...

Some data providers, for example ORATS, have snapshot data of at 3:46 PM, ie. 14 minutes before market close. That's the right way to do.
 
Last edited:
Any feedback on the best time to sample the bid/ask? Some vendors seem to do that immediately (ie seconds) prior to the close, whereas others are 15/30 mins prior to close...
 
You must log in or register to reply here.
Back
Top