# How to estimate market risk using only publicly available data?

How can I calculate market risk for the US Stock Market (NYSE or NASDAQ) using only freely accessible data? I'm only interested in the market risk of the whole economy not of different industries, companies or sub-sectors.

Somebody suggested to me to download market index data and construct the volatility as a measure of market risk. But I can't find a proper explanation how I do this.

Thank you.

p.s. Is the following procedure correct?:

I get data US Stock market (for example https://fred.stlouisfed.org/series/SP500) calculate the growth rate and in the last step I compute the variance of the growth rate to get the market risk?

• There is not one canonical definition of market risk and many to chose from. You seem to have no argument to prefer one in particular. In this situation, I would just choose the definition of market risk that is easiest to compute with the data you have. Let's consider the risk that the market does not open tomorrow. That's certainly a risk. This can be computed using exchange calenders (available online for free) past data (of the sort you found) and a simple probability model – user189035 Mar 26 '18 at 12:54
• Thank you for your comment. But I was thinking more about the method used by Campbell et al. 2001 (onlinelibrary.wiley.com/doi/full/10.1111/0022-1082.00318) – PAS Mar 26 '18 at 14:17
• Even then they detail more than one measure of market risk. You'd have to be more specific. – user189035 Mar 26 '18 at 15:00
• They define market risk as the the variance of the average return of the stock at date $t$ and the average return over the whole sample from 1962-1997. Its formula 17 the values are depicted at page 11. – PAS Mar 26 '18 at 17:43