Mahalanobis Distance

We are big fans on using distance measures while prospecting for investment strategies. Previously:

Recently, we came across an interesting paper, Skulls, Financial Turbulence, and Risk Management, Mark Kritzman, CFA, and Yuanzhen Li, that uses the Mahalanobis distance to construct a turbulence index. The basic idea is that the more asset returns break from the past, the more “significant” a market event.

We took the basic intuition behind this and constructed a portfolio that switches between equities and bonds based on the Mahalanobis distance between them.

The out-of-sample results, factoring in transaction costs, look promising but doesn’t really stand out compared to other, more dumber, strategies that avoid steep drawdowns. However, two points over the Midcap buy & hold cannot be dismissed outright.

The code, charts and paper are on github.

Large Moves Happen Together

We are often told that missing the 10-biggest days in the market leads to sub-par returns. While it is certainly true, what is often not said is that those really big days occur around really bad days. Welcome to tail-risk.

The average daily return of the NIFTY 50 is 0.06%. The worst daily return is ~ -13% and the best is ~18%. Welcome to tail-risk.

In any given year, there are a lot of days when returns fall out of 1, 2 or even 3 standard deviations (σs), Up and Down.

And these σ moves tend to happen close to each other. i.e., volatility clusters.

What the histogram above is showing is that most of the 3σ moves have happened within 5 days of each other! Let’s zoom in on a 10-year period of the index and mark the outliers on it:

Now, lets pick a very simple actively managed strategy that tries to side-step the σ moves. The details of the strategy itself is unimportant. Suffice to say that it creates excess returns compared to buy & hold.

The average daily return of this strategy is 0.07%. The worst daily return is ~ -7% and the best is ~18%. At least some of the left-tail has been clipped but at what cost?

Notice how both the number of large Up and Down days are lower here compared to buy & hold?

Outliers still cluster but there a lot less of them.

This is the nature of market volatility. Investors have to either commit to buy & hold and catch all the moves or commit to an actively managed strategy knowing that while trying to side-step σ moves, some +σ moves will also be sacrificed. It is the FOMO that keeps investors switching between the two, resulting in sub-par returns.

Options Weekly 04.03.2023

Summary: Mar NIFTY 18000 calls added 65,14,850 contracts while 17400 calls shed 44,21,500. On the Put side of the equation, the 17500 strike added 80,10,000 while the 16850’s shed 4,60,800.

MAR NIFTY OI

MAR NIFTY OI chart

MAR BANKNIFTY OI

MAR BANKNIFTY OI chart

MAR NIFTY Volatility

MAR NIFTY Volatility chart

MAR BANKNIFTY Volatility

MAR BANKNIFTY Volatility chart
Dotted lines indicated actual underlying volatility. Solid lines are IVs.

VIX Density Plot

VIX kernel density plot

Nifty 50 Returns Density Plot

NIFTY 50 kernel density plot

Sharpe vs. Rates

Sharpe Ratios are often used to sort through competing investments. It is the original “risk adjusted returns.” It’s a mathematical expression of the insight that excess returns over a period of time may signify more volatility and risk, rather than investing skill (investopedia, wikipedia).

Rb or Rf, the risk-free return, is usually cumbersome to handle. So, typically, it is either set to zero or a constant value. The problem is that rates vary over time and has an impact on the relative ordering of investments.

At high interest rates, SR(mid-caps) > SR(large-caps)

Ideally, you want your Sharpes to be positive and stable. Unfortunately, that is never the case.

During bull markets, Sharpes trend up and reverse course in bears.

At the end of the day, it boils down to whether the returns are worth the risk.

You might very well be trying to catch lightening in a bottle.

For more index stats, visit our Index Metrics dashboard.