Rocket Science? Meet Quant Trading.

Quant Galore
5 min readAug 28, 2023

A wise man once said that trading isn’t rocket science. Well… sometimes it is, literally.

To those outside of finance, what quantitative traders and hedge funds do may seem like rocket science. However, although what really goes on behind the scenes may not be too complex, sometimes, it really is rocket science.

So, today, we’ll be taking a dive into how Wall Street quants apply actual rocket science concepts to real-world trading.

Signals Are Everything

While Aerospace Engineering (rocket science) and Finance may not have much in common, the biggest similarity is that they both use tons of data. In both fields, most of this data is in the form of a time-series, where the data points are observations across time:

Source
Source

In both fields, the data is used to generate insights, but most importantly, it’s used to influence decision making. Making data-driven decisions are especially important when it comes to rockets, as errors in performance can lead to deadly and costly mistakes.

Because of this, engineers have worked tirelessly to create a framework for getting the absolute most of their data:

Digital Filtering

A digital filter is essentially just a way transforming data to make it more interpretable. This transformation is done to identify trends and better spot anomalies. To see why this is necessary, let’s look at how one dataset can look very different after filtering:

Original Data:

Source

A common type of filtering is low-pass filtering. This essentially just smooths the data so that fast, rapid changes don’t alter the message of the data:

Low-Pass Filtered (Smoothed):

This is done to identify long-term trends and to identify baseline average values.

On the opposite end of low-pass filtering is high-pass filtering. Instead of smoothing out the data, this extracts only the fast, rapid changes:

High-Pass Filtered

To see why digital filtering is important, let’s walk through an example from both rocket science and finance.

In Space:

You are a pilot for a small rocket company and have made your first venture into space. You’ve just launched, so you can’t sit back on auto-pilot just yet and have to monitor and adapt to the signals on your screen. Most launch failures happen due to combustion issues, so you navigate to the screen which features the engine data.

Generally, this process will be fully automated since the changes need to happen quickly and precisely, but here’s what might happen:

  1. The system may combine multiple measures into 1 low-pass smoothed signal that represents overall combustion health (e.g., temperature, + fuel + oxygen level, etc.)
  2. If that signal then dips to levels below its long-term average, the signal will be filtered into each of its high-pass components.
  3. If the deviation is caused by the temperature signal, the system will activate the cooling systems to get the combustion health back on track.

In Finance:

Since finance has recruited from aerospace talent for decades now, there has been ample time to integrate these filtering techniques into almost every aspect of trading — some of which you may be well familiar with.

High-Pass

There’s perhaps no better example of high-pass filtering than volatility data:

While volatility itself isn’t a filter, it acts as one by highlighting the short, fleeting, and rapid changes to the underlying data — filtering out the smoother, slower changes.

In practice, this is mostly used as the basis for the timing of market entries, but by extrapolating it into an option selling strategy, it can be extremely profitable: Volatility Trading is Back, Seriously.

Low-Pass

On the other end, there’s perhaps no better example of low-pass filtering than moving averages:

By smoothing out all of the rapid, short-term prices, moving averages help to act as a gauge of long-term trends and to better answer the question of “how far is this data point away from the average we expect?”

In practice, this is sometimes the starting of many factors used in investment decisions. For example, if today’s technology sector prices are 3 times greater than the 200 day moving average, a contrarian investor may consider shorting the sector since they deem it as overvalued.

So overall, while Finance generally isn’t as tough as rocket science, it’s done a terrific job of inheriting some of the fields common practices.

If this article piqued your interest, you’d definitely enjoy some of my other posts just like this one:

Happy trading! 😄

--

--

Quant Galore

Finance, Math, and Code. Why settle for less? @ The Quant's Playbook on Substack