Estimating Execution Prices from Spot, Size, and Bid–Ask Spreads
Summary
The document asks how to estimate an execution or settlement price from a spot price when order size affects the realized price. It proposes using the bid–ask spread as a starting point because both spread and execution impact depend on liquidity, and asks whether a linear, quadratic, cubic, or other size adjustment would be appropriate. Its illustrative examples show a larger percentage premium over spot for a larger transaction, suggesting that execution impact may not scale proportionally with quantity.
The document does not present a fitted model, study, or empirical evidence supporting a particular functional form. It is a practical modeling question motivated by a system that cannot submit limit or stop orders, so the estimate would need to be made before execution. The examples are hypothetical and do not establish a general relationship across commodities or liquidity conditions. Any usable estimate would need instrument-specific data on executed prices, order sizes, and market conditions; the proposed spot-and-spread relationship alone is not validated here.
Key ideas
- Execution price can differ from spot, and the difference may depend on order size and liquidity.
- The document proposes the bid–ask spread as a possible input to an execution-price estimate.
- Its hypothetical examples suggest that percentage price impact could rise with transaction size.
- It provides no evidence for choosing a linear, polynomial, or other size-impact function.
Tags
Full text
# Settlement/Spot/(bid ask spread) ratio # Settlement/Spot/(bid ask spread) ratio Are there any studies on the average difference or ratio between Settlement (execution price) and the Spot price dependant on lot size. I'm looking for a function such as `SETTLEMENT=SPOT*LOT_SIZE^CURVING_FACTOR/commodity_constant*ASK/BID`. This settlement/spot ratio is dependent on the amount of liquidity and the commodity type. However the ask/bid spread is also dependent on the amount of liquidity. So I was hoping that the ASK/BID spread would be a good starting point. However I don't have enough knowledge to decide what the best "curving factor" I should use, quadratic/cubic/exponentioal/linear... Is this the correct way to model it? I have to do this because I'll be working with a system that can't do limit/stop orders and so I need to estimate what the execution price will be before execution. ## Example A 10 quantity lot transaction with a spot of 100 USD/lot could yield a settlement after execution of 1002 USD. This is a .2% increase over the spot rate/lot. However a 100 quantity lot transaction could yield a settlement of 1010 USD giving a 1% increase of the spot.
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.