Tuesday, December 27, 2011

Risk indicators

During the financial crisis of 2008, I wrote about how I watched some risk indicators such as the VIX or the TED spread  to decide what leverage I should use for my trading strategies. It turns out that this procedure is just as critical for the current crisis that began in August 2011. In fact, more than leverage-determinants, they can be used as the all-important variable that determines whether a certain strategy should be run at all. (What's the point of running a model that you think will lose money with low leverage?)

There are now more than a few of these risk indicators to pick from. Besides the VIX and the TED, there are the VSTOXX (EURO STOXX 50 Volatility), the VXY (JPMorgan G7 Volatility Index), the EM-VXY (JPMorgan Emerging Market Volatility Index), the ETF's ONN and OFF, and probably many more that I haven't heard of yet. 

A lot of academic research has been done on whether we can devise "regime switching" models based on some complicated pattern-recognition algorithms to decide whether a market is in a certain "regime" which favors this or that particular model or parameter set. And often, these regime switching models rely on the recognition of some complicated set of patterns in the historical price series. Sorry to say, I have not found any of these complex regime switching model to have any real out-of-sample predictive power. On the other hand, my research shows that some of the aforementioned simple risk indicators will indeed prevent some trading models from falling off the cliff.

But which of these indicators are applicable to which model? This is not so obvious. For example, you might think that the EM-VXY would be an ideal leading indicator for Forex trading models that involve emerging market currencies, but I have found that it is only a contemporaneous (and thus useless) indicator to mine. Another example, I said during the 2008 financial crisis that VIX seems to be a useless contemporaneous indicator for equities trading models, but strangely, it is a good leading indicator for FX models. In contrast, the TED spread that everyone were obsessed about in 2008 shot up to over 300 bps then, but never went beyond 100 bps this time around. So really only rigorous backtesting can guide us here.

What risk indicators do you use? And have you really backtested their efficacies? Your comments would be very welcome here.


Anonymous said...

Hi Ernie.

First, I appreciate your blog, lots of valuable thoughts here.

It´s not exactly what you talked about but I backtested a daily CAD/CHF momentum strat with the S&P500 to see if the latter shows any predictive/leading caractheristic regarding the signals generated by this strategy (both long and short).

Turned out that to some extent yes, it would exclude more false signals than winning ones if it was used as a parameter in this case. I think if it increases profitability then it might make sense to add, though I still want to better understand which instrument leads and can predict price action in the other. I.e. if both instruments turn exacly one day before the signal is generated then it is not clear. (note it´s a daily strategy)

Great if you can give your view.

Anonymous said...

Sorry, the last sentence goes:

Great if you can give your view on the above and on how you backtest leading/contemporaneous characteristics regarding two instruments.

Mr Blaster said...

Hi Ernie,
This is a very interesting post, thanks for sharing.

I am a big believer in price based systems. I have & I saw some price based only system, both EOD & intraday, work consistently very well with more than hundreds(EOD) & thoudsands(Intraday)of trades.

Though I have posted & doubted that the whether the "risk" is due to the uncertainty of price see the post...


...but I use price based historical volatility as my risk indicator for signaling bear/bull market which shows no false signal in singaling bear market for more than 20 years & more than 1 market.

What I want to elaborate is actually I use voaltility as a regime indicator. I am a chatist too. I do have some sujective chart reading rules, like "high" & "low" volatility regime.

Thanks again!! It is an inspiring post.

Storugglan said...

Nice blog and post! What do You think abput using Williams VIX Fix? Is it out of question?

Ernie Chan said...

Hi src,
I believe what you are saying is that SPX is a leading indicator of CADCHF, but you are not sure whether it leads CAD or CHF? I think you can discern by testing it against USDCAD and CHFUSD separately. But I am not even sure that you have to find out to make this strategy better.

Ernie Chan said...

A typical way to test for leading correlation is just to test for any correlation between a returns from one price series and the lagged returns from another price series.

Ernie Chan said...

Market Blaster,
Thanks for the link to your post. But the conclusion of your post seems obvious: if you assume that the price series is stationary (i.e. with a fixed stddev), then of course any mean-reverting strategy will be profitable (just buy low and sell high).

I am, however, intrigued to hear that you find historical volatility to be a good risk indicator!


Ernie Chan said...

Hi Storugglan,
The Williams VIX Fix does seem like an interesting replacement for VIX. I will do some research to see if it corroborate with the claims in the article.
Thanks for the idea!

Bernd said...

Hi Ernie:

Could you please develop a little bit further about the August 2011 crisis you speak about?

DavidD said...

The Baltic Dry Index, measures the cost of freight. Not so much a proxy for risk,but useful to gauge if any market improvements are due to increased trade or accountancy tricks.

Ernie Chan said...

Hi Bernd,
The morphology of the crisis that starts in August 2011 can be most easily seen from a 2-year chart of VIX. You can see that VIX had been below 30 for almost 1 year until August, when it shot up to over 45 at one point. The initial cause of this volatility is the S&P downgrade of the US federal debt, but then investors became much more concerned about European sovereign debt. This volatility and news-driven environment wrecked chaos with many trading strategies, leading to one of the worst years for the average hedge fund performance.

Ernie Chan said...

Yes, that's a good macroeconomic measure that an ex-partner of mine is very fond of.

The Imperatore said...

Hi I watch the overnight funding rates, eurodollars futures and currencies, mainly EUR/JPY and AUD/JPY



farmland investments said...

I am not an expert trader by any means, but I basically just use the VIX. I own a large stock portfolio, and hedge it with the iPath Short-Term VIX ETF.

Anonymous said...

If your proxi of risk is VIX, maybe usefull to observ also Skew Index of CBOE.
Looking at past, often we have had periods of low VIX joint at high SKEW, a very dangerous mix using some form of Kelly's leverage.
Reading pobabilities table written in CBOE SKEW White Paper, it seems that VIX=30 and SKEW=110 is a bit less risky than VIX=20 and SKEW=145 for events over 2 SD : so, if you use today VIX to forecast tomorrow's variance in Kelly formula, maybe usefull adjust VIX level in some arbitrary form NEWVIX=f(VIX,SKEW).
NEWVIX=VIX*(0.0192*SKEW - 0.9094) works fine for me.


Ernie Chan said...

Thanks for giving us a handy adjusted VIX formula for outliers. Very interesting.

jack leon said...

Really helpful article , thanks Ernie and other people who commented here

Slim said...

It's tough for me to believe in any one of these risk indicators. I feel like it's almost more of a judgement call rather than a quantitative measure. For example, the VIX, it's mostly useless as a forewarning of risky times in the near future. It's inversely correlated with the equity markets. Markets rise and VIX typically falls, markets falls and VIX typically rises. Markets tank and VIX pops like a mother... Thanks for letting me know AFTER a 5% decline in the equity markets that volatile times are ahead. Almost all of these indicators are like that. And all major asset classes are so correlated nowadays anyway. Nothing goes volatile without everything else following.

Take a look at the period right before the euro crisis. Indicators won't tell you much. But there was the debt ceiling debates, potential US credit downgrade, Euro uncertainty, etc. Non quantitative measures that foreshadowed a high chance of rough times ahead. Of course there was no measure that could have predicted the Japan nuclear reactor debacle.

If high vol periods are seriously detrimental to your strategy, why not figure out a way to hedge against it? Buy vol as insurance? That kind of insurance is usually pretty expensive though...


Ernie Chan said...

Yes, conventional wisdom holds that VIX is a contemporaneous indicator. However, my backtests show that it does have predictive value. Also, long-dated options may be more predictive value than the front month contracts. From what I hear, those long-dated options are currently predicting a higher volatility than the front months.

Penryl III said...

I use BDIY for trading but mostly don't use any. I usually look forHold Long period.