Author: shyam

VVIX for VIX

Came across an intriguing post by Markku Kurtti: Employing volatility of volatility in long-term volatility forecasts (outcastbeta). The gist of it is that the volatility of volatility and average volatility is more predictable. So, predict them separately and then put them together to get a volatility forecast.

Our previous attempts at forecasting VIX have led us the conclude that nothing beats locf. Just extending the last value of VIX forward has beaten all of the models we’ve looked at.

What if, we applied the process outlined in Mr. Kurtti’s blog for VIX?

Not looking so good if you forecast out for 20-days.

A good forecasting model should be able to estimate large spikes and reversion to mean from them. If not the former, then at least the latter. So, if a model is good at the mean-reverting bit, then you can combine both locf and the model to get better predictions.

For example, if you only consider the 1-day forecast, then obviously locf has the upper hand.

However, as you try to predict farther away, you expect the model to nail the mean-reversion bit.

That is not the case here.

Once again, locf wins.

Code and charts up on github.

Categories Tags ,

CGMM for VIX

Previously, we explored using Meta’s Prophet library to predict VIX. It turned out that simply extrapolating the last value of the index worked better (Prophet for VIX). In fact, locf (last one carried forward) works better than GARCH(1, 1) and most other approaches. Can the same be said about Conditional Gaussian Mixture Models?

We used the cgmm python library to forecast 20-day forward VIX and compared its root-mean-squared errors (rmse) to those of locf‘s.

locf is pretty hard to beat.

Especially so when the VIX index itself is volatile.

Code up on github.

Categories Tags ,

Leverage

When you buy a stock, you buy a piece of business, not a quote from a broker. – Naved Abdali

There was a time when people actually believed that buying a stock meant that you were buying a piece of the underlying business. However, we have strayed far, far away from that. The average holding period of equities keeps dropping and everything has become trading sardines.

If we are all traders anyway, why not trade with leverage? Or so goes the clarion call of brokers offering the Margin Trading Facility.

Here’s how it works:

Initial Margin: You pay a portion of the total trade value from your own funds.
Broker Funding: Your broker provides the remaining amount to complete the purchase.
Collateral: The stocks you buy are pledged as collateral to the broker.
Interest: The broker charges daily interest on the borrowed funds.

Margin Trading Facility (MTF) has become a more significant revenue stream than traditional brokerage fees. And the market has been expanding.

Judicious application of leverage, especially on low-volatility portfolios, does make sense. However, the ground reality is different. As a percentage of free-float capitalization, its mid/small-cap momentum stocks that are being financed on margin.

For example, back in August 2024, almost 75% of Motilal Oswal Nifty Smallcap 250 ETF’s free-float capitalization was financed by MTF. If a situation were to have arisen where most of the traders who had bought this on leverage couldn’t meet a margin call, the broker’s risk management team would’ve just dumped this in the market. Given the already abysmal liquidity situation of that ETF, it would have been enough to trigger a cascade of stop-losses, further dislocating the market.


As it turned out, the index would proceed to go down 25% by March 2025.

Currently, more than 15% of Nazara – a company who’s entire business model has been legislated out of existence – is financed under MTF. I hope traders know that MTF is a full recourse loan?

As much as leverage exposes you to risk, there’s another side to this equation. Keeping track of the most leveraged stocks can help you identify margin liquidations and profit from it. However, the pain of liquidations will be felt by many but the profit will go to the few.

Categories

Book Review: Apple in China

In Apple in China – The Capture of the World’s Greatest Company (Amazon,) Patrick McGee lays out the how, what and why of Apple’s outsourcing efforts in China.

The growth of the manufacturing ecosystem in China resembles the growth of most social networks. First comes the hook: the “wow” factor. Then come the key nodes: the influencers. Before you know it, half your friends are there and you are compelled to join them or be left out. Then comes the theft of property by the owners of the social network (your photos, your activity, your thoughts, etc. are all theirs.) And then the monetization.

Steve Jobs initially wanted all of Apple’s products to be made in America. He was of the opinion that it was the only way in which he could maintain the strict standards that went into making Apple’s products. How did they end up making everything in China?

At first, the hook: Terry Gou of Foxconn.

It reached a point where Apple (the influencer) was co-developing advanced manufacturing processes with vendors in China.

Slowly, and then suddenly, all of component manufacturing shifted to China.

Then, the theft started.

Leading to the final step: monetization.

Apple finds itself enmeshed in China and scaling India in is proving difficult.

Primarily because of language politics, unionization and the perennial election cycle that prevents the Indian government from doing what China did: suppress worker rights.

Facebook did not “die” in the traditional sense, like how MySpace did. It just became unpopular with the next set of cohorts. It is aging out, continuing to monetize its current user base, while Instagram attracts the next generation. A similar thing might happen to the Chinese manufacturing juggernaut because now, everybody know that:

In China, ‘win-win’ means China wins twice.

Categories Tags ,

The Smirk, Part II

While the concept of volatility smirk is simple, the pattern itself is unstable. For example, different expiries have different shapes.

And these shapes change across days as well.

One way to keep track of these changes is by fitting a model through the implied volatilities. Here, we fit a parabola (y = ax2 + bx + c). a, the coefficient of strike_pct2, gives a measure of the narrowness/steepness of the smirk.

By sampling the curve and tracking these coefficients, you can begin to form an opinion on what is “normal” vs. a trading opportunity.

Code and charts on github.

Categories Tags ,