Interpreting Order Book Price–Size Correlations with Realized Volatility

Summary

The document raises a market microstructure question about how correlations between bid or ask prices, opposite-side quoted sizes, and realized volatility should be interpreted across a sample of stocks. The author expects larger quoted size alongside a high opposing quote to be associated with greater price movement, but reports that the plotted correlations appear negative for most stocks. The text frames this as a question about whether the observed direction makes sense, rather than presenting a resolved finding.

The discussion is limited: the referenced graph is not included in the text, and the realized-volatility formula is introduced but not supplied. It gives no sampling interval, quote aggregation method, volatility horizon, or analysis of whether correlations are contemporaneous or lagged. Consequently, the proposed intuition cannot be evaluated from the document alone, and correlation does not by itself show that quote sizes cause volatility. The useful takeaway is that order book relationships require careful definition of quote variables, timing, and volatility measurement before interpreting their sign.

Key ideas

• The document asks how quote prices and opposite-side sizes relate to realized volatility across stocks.
• The observed correlations are described as negative for most stocks, contrary to the author’s intuition.
• The text does not include the referenced graph or the stated volatility formula.
• Correlation patterns alone do not establish that order book quotes cause volatility.
• Interpretation depends on the timing and construction of quote and volatility measures.

Tags

# correlation between bid/ask prices and bid/ask sizes to realized volatility


# correlation between bid/ask prices and bid/ask sizes to realized volatility












I am trying to understand the effect of correlation between bid price and ask size with realized volatility (called targe_vol). Similarly, correlation between ask price and bid size with realized volatility. This is shown in the graph below for data of 112 stocks. Does this picture make sense? When the bid size and ask price are both high then price will have to move towards ask price thereby increasing volatitliy. similarly, when bid size is high/(low) and ask price is small/(large) then price move is smaller/(no trade happens) and real. vol. should be smaller too. By this logic, I should see a positive correlation of bid price and ask size correlation to real. vol. Why is this inverted?

There seems to be a general pattern here because it's negatively correlated for most stocks.

The formula for realized volatility is as follows:

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.