Modeling Nikkei 225 Futures Slippage with Volatility-Based Spread Assumptions

Summary

The document considers how to represent execution costs when backtesting a Nikkei 225 futures strategy that uses market orders. It explains that slippage depends on conditions such as volatility and trade size, so a single fixed cost may not describe execution well. It proposes a rough first-pass model that assigns wider assumed spreads to higher volatility ranges, using volatility index levels as the conditioning variable.

The suggested spread values are explicitly illustrative, not measured estimates for a particular contract, exchange, broker, or order size. The response gives no transaction data or validation method, and notes that building a realistic estimate would require substantial work. Traders can treat the idea as a starting point for scenario analysis, then refine it with observed fills, fees, and market conditions before relying on backtest results.

Key ideas

• Slippage varies with volatility and order size.
• A volatility-conditioned spread assumption can provide a rough first-pass execution model.
• Illustrative spread values should not be treated as empirically validated costs.
• Realistic backtests need cost estimates suited to the contract, venue, broker, and trade size.

Tags

# Answer by user42108 (score 1)


# When backtesting Nikkei225 futures with market orders, how many points to account for eventual slippage and trading costs?












I want to backtest a strategy based on Nikkei 225 futures (preferable at the Singapore exchange).

I am using market orders for entry and exit.

Although I now that theoretically market orders for a very liquid instrument should not have any slippage, I have heard that sometimes one does not get executed immediately or other strange things happen.

What is a reasonable amount of points or amount of money to account for slippage and costs for exchange and broker?

## Answer by user42108 (score 1)

https://quant.stackexchange.com/a/59441

What is a reasonable amount of points or amount of money to account for slippage and costs for exchange and broker?

Slippage will depend on many things - volatility and size are probably the most important. Your question is non-trivial and trying to get a realistic answer would be a lot of work. As a first pass, I might try a toy model that assumes, for e.g., that the spread is 5 points if VNKY is <20, 15 points for 20-30 and 25 points for >30 (make up your own numbers).

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.