Skip to content
All library documents

Using Order Book and Trade Imbalance to Assess Short-Term Pressure

Article Quant Q&A · Author: David
Read the original

Summary

The document considers two measures of near-term buying and selling pressure for a market-making strategy: comparing displayed volume across several bid and ask levels, and comparing recent buyer-initiated and seller-initiated trades. These are identified as order book imbalance and trade imbalance, respectively. Both can help with short-term price discovery, but the post does not establish that either measure reliably predicts price movement or execution opportunities.

The key design choice is the observation window. A fixed number of order book levels may reach prices that are unlikely to trade under current volatility, while a fixed number of transactions can represent different amounts of elapsed time and market activity. Depth and trade-count parameters should therefore reflect the strategy’s intended horizon. The discussion offers a starting point for measurement, not evidence of profitability or a complete market-making system.

Key ideas

  • Displayed bid and ask volume can be compared as an order book imbalance measure.
  • Recent buy and sell transactions can be compared as a trade imbalance measure.
  • Order book depth should be chosen with price distance and current volatility in mind.
  • The number of trades included should fit the intended short-term observation horizon.
  • Imbalance measures are starting points for price discovery, not guaranteed signals.

Tags

EquitiesMarket makingMarket microstructureExecution
Full text
# Checking short term supply and demand in the stock market


# Checking short term supply and demand in the stock market












I am implementing a pretty simple market making strategy. I want to see if the demand is higher than the supply in the short term so that I will be able to buy and sell decently fast. My goal is to be able to buy and sell within the same minute, if certain market conditions are met.

I've thought about checking the order book and looking at the first 5-10 prices on the buy and on the sell side and then compare them. If there is more volume on the buy side than volume on the selling side, then I can assume that there is more demand than supply currently on the market.

Then I will check the time and sales and see if there were more transactions to buy than transactions to sell in the past 30-40 transactions.

Is it correct to think about short term supply and demand this way ? I'm using a market making strategy in a mean-reverting highly liquid market.

## Answer by quantinho (score 1, accepted)

https://quant.stackexchange.com/a/77063

Well, the first approach you mentioned is called orderbook imbalance while the second is trade imbalance. It is hard to say whether those are "correct" or "incorrect" approaches because it all depends. However, those two approaches can be used as price discovery.

In the first approach, I would ask why 5-10 levels? How far is the price at depth 10? If volatility is low and depth N price is unlikely, than it is not a good measure of short term measure of supply/demand. Depth parameters should reflect what short term means for you.

Same can be said about the number of transactions. Overall, orderbook/trade imbalance is a good starting point.

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.