Yesterday's mini crash in a historical context. Jon Danielsson. Modelsandrisk.org

Yesterday's mini crash in a historical context

February 6, 2018
The stock market had a mini crash yesterday. So how big was that in a historical context?

If I take the entire history of daily S&P 500 returns, which I have from 1929, and plot it on a log scale, you can't really see what happened yesterday.

Using volatility to predict large losses on the SP-500 compared to actual losses

By zooming into this decade, it becomes more clear, and it does not seem to be all that significant.

Using volatility to predict large losses on the SP-500 compared to actual losses

However, still hard to see. So why not look at returns (actually log returns), and for good measure draw a red line through the returns at the magnitude of yesterday's crash. Plenty of worse crashes it seems.

Using volatility to predict large losses on the SP-500 compared to actual losses

By zooming into this decade, can see that we haven't seen one like it for a few years.

Using volatility to predict large losses on the SP-500 compared to actual losses

So how big was the crash in a historical context? Let's look at the biggest drops in my history, and plot them sorted, identifying the two biggest as well as yesterday:

Using volatility to predict large losses on the SP-500 compared to actual losses

So it was the 127th worst drop. The history is 88 years, so the magnitude of yesterday was a once in an eight month event.

Not really an attack by a dragon was it..


Artificial intelligence and the stability of markets
Cryptocurrencies don't make sense

Models and risk
Bloggs and appendices on artificial intelligence, financial crises, systemic risk, financial risk, models, regulations, financial policy, cryptocurrencies and related topics
© All rights reserved, Jon Danielsson,