Skip to content
All library documents

When Stochastic Calculus Matters in Practical Trading

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

Summary

The discussion distinguishes trading for exposure from quantitative work that develops valuation models. A practitioner may use outputs from stochastic-calculus-based models, such as Black–Scholes, without applying stochastic calculus directly to routine asset trades. The answer suggests that specialized product and derivatives roles at banks are more likely to require the mathematics for building and assessing pricing models.

Stochastic calculus may also matter to a trading strategy when a trader believes a prevailing model misprices an instrument and sees a possible arbitrage. The answer presents this as a conceptual explanation, not an empirical study or a universal rule. It gives no test for when a model discrepancy is exploitable and does not address other trading applications, such as risk analysis or dynamic hedging, in detail. Its claims are explicitly speculative and depend on the trader’s role and instruments.

Key ideas

  • Routine trading for asset exposure may not require direct use of stochastic calculus.
  • Derivatives product quants may need stochastic calculus to build valuation models.
  • Model discrepancies can motivate a search for arbitrage opportunities.
  • The answer is a role-based perspective, not an exhaustive account of applications.

Tags

Derivatives pricingArbitrageOptionsStatistics
Full text
# What is the role of stochastic calculus in day-to-day trading?


# What is the role of stochastic calculus in day-to-day trading?












I work with practical, day-to-day trading: just making money. One of my small clients recently hired a smart, new MFE. We discussed potential trading strategies for a long time. Finally, he expressed surprise that I never mentioned (much less used) stochastic calculus, which he spent many long hours studying in his MFE program. I use the products of stochastic calculus (e.g., the Black-Scholes equation) but not the calculus itself.

Now I am wondering, does stochastic calculus play a role in day-to-day trading strategies? Am I under-utilizing a potentially valuable tool?

If this client was a Wall Street investment bank that was making markets in complicated derivatives, I'm sure their research department would use stochastic calculus for modeling. But they're not, so I'm not sure how we would use stochastic calculus.

(Full disclosure: I have Masters degrees but not a PhD. I'm an applied mathematician, not a theoretician.)

## Answer by Shane (score 25, accepted)

https://quant.stackexchange.com/a/153

This is pure speculation:

MFE's are really tailored toward valuation models (how can we develop a model to price x swap, etc.). You don't entirely have to worry about those details in order to trade them: you're just quoted a price based on these models. But if you go in-house at a bank and are working as a product quant (structured products, etc.), then you really need to worry about these things.

Alternatively, it could be relevant to a trading strategy if you think that the current model is mispricing things and there's an arbitrage opportunity. This is why banks have put so much effort into having good models, and jump at opportunities for very minor improvements. This kind of behavior is documented in Derman's "My Life as a Quant".

Short of that, if you are simply trading an asset in order to gain a specific kind of exposure, stochastic calculus is not really used very much.

As a final note, I would point to the draft of Steven Shreve's "Stochastic Calculus and Finance" as a free reference, if you're looking for one.

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.