Welcome to MeatPy Discussions!

[vgreg]

👋 Welcome!

We’re using Discussions as a place to connect with other members of our community. We hope that you:

• Ask questions you’re wondering about.
• Share ideas.
• Engage with other community members.
• Welcome others and are open-minded. Remember that this is a community we
  build together 💪.

To get started, comment below with an introduction of yourself and tell us about what you do with this community.

[rumcode]

Would it be possible to use the parquet file as a staring point to examine the cancellations, orders etc at specific time interval. My idea is that to use the parquet file limit order book to jump to a specific point in time in the gz file. For example suppose I want to see the activity in the limit order book starting ta 10:00. The parquet file gives me the state of the limit order book at that time. Thus any additions, deletions, fills starting at 10:00 could be reviewed by racing through without creating the lob until 10:00 and then message by message at that time.

That is, I'm trying to find a way to speed up the process to look at the dynamics of the limit order book at an interval in time.

From a research stand point I want to look at the dynamics of order cancellation as most researchers who use TAQ fail to incorporate order cancellation into their understanding of limit order book dynamics. So as a first crack, I just want to look the message/order book dynamics for a period of time.

[vgreg]

I think this is technically possible, but it would require rewriting a lot of code. When you say "examine the cancellations .. at a specific time interval", do you mean doing it interactively, or do you mean recording events and order book only for a specific interval?

The most costly operation is recording events and storing them to disk, so if you want to record only a specific interval, you can set record_start and record_end for each recorder that you are interested in. It will still have to replay the full order book, but it will avoid any write operation until the interval of interest.

MeatPy/src/meatpy/event_handlers/lob_event_recorder.py

Lines 26 to 36 in 7c0c641

	class LOBEventRecorder(MarketEventHandler):
	"""Records limit order book events and manages data output.
	Attributes:
	records: List of recorded measures
	record_timestamps: Optional list of timestamps to record
	record_start: Optional start timestamp for recording
	record_end: Optional end timestamp for recording
	write_csv_during_recording: Whether to write CSV during recording
	output_file_name: Output file name for concurrent writing
	buffer_size: Size of the buffer for records

[rumcode]

Thank you for your reply. When I read through the academic articles, they seem to be missing the importance of hidden orders or those which are cancelled. Market makers like to entice the buyer to "reach for it" in order to fill his order like Kung Fu Panda reaching for his meal of dumplings.

…

________________________________ From: Vincent Grégoire ***@***.***> Sent: Wednesday, December 3, 2025 6:13 AM To: vgreg/MeatPy ***@***.***> Cc: rumcode ***@***.***>; Comment ***@***.***> Subject: Re: [vgreg/MeatPy] Welcome to MeatPy Discussions! (Discussion #59) I think this is technically possible, but it would require rewriting a lot of code. When you say "examine the cancellations .. at a specific time interval", do you mean doing it interactively, or do you mean recording events and order book only for a specific interval? The most costly operation is recording events and storing them to disk, so if you want to record only a specific interval, you can set record_start and record_end for each recorder that you are interested in. It will still have to replay the full order book, but it will avoid any write operation until the interval of interest. https://github.com/vgreg/MeatPy/blob/7c0c641c029370620193fa344c73e413f0fa2ae1/src/meatpy/event_handlers/lob_event_recorder.py#L26-L36 — Reply to this email directly, view it on GitHub<#59 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ADHLVRIWZINSIGILCAFJV7D373HWTAVCNFSM6AAAAACCJWDVHWVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTKMJUHAYTIMI>. You are receiving this because you commented.Message ID: ***@***.***>

[vgreg]

It is true that these features are often overlooked, but some articles did look at them.

For hidden orders for example see https://www.sciencedirect.com/science/article/pii/S1386418122000313

Even in one of my paper we look at hidden orders, albeit in a specific context (after-hours around earnings): https://onlinelibrary.wiley.com/doi/pdf/10.1111/1475-679X.12394

As for order cancellations, I don't remember any paper along the lines of what you suggest, but they have been looked at in many papers such as https://onlinelibrary.wiley.com/doi/pdf/10.1111/jofi.12769 and https://academic.oup.com/jfec/article-pdf/12/1/47/2439285/nbt003.pdf