VVV Weight Calculations: Prepared for the Downside and Primed for the Upside
Article 3 of 7
- Two of the most essential ingredients in determining weights are volatilities and variances (also covariances) of assets.
- In the “Velocity of Volatility and Variance” (or VVV) crash protection mechanism, we adjust the volatilities and the variances (including covariances) of assets depending on how fast they are likely to change during market crashes.
- Using VVV, our portfolio can outperform, in terms of returns, typical portfolios using more conventional weighing mechanisms by almost 80% with a considerably higher sharpe ratio (e.g. OURS: 1.91 vs OTHERS: 1.44). We achieve this by taking slightly higher risk and based on our belief that volatility is a small price to pay for the convenience of trading anything from anywhere and anytime, as long as we are sufficiently equipped to deal with downward movements.
The third article, of seven planned articles in our Eiffel release plan series, will provide a summary of our asset weight calculations. An earlier article has a very detailed explanation of our methodology: [LINK]. Our novel portfolio weighting technique considers certain stylized facts about the financial markets. We have then tweaked the weight computations to factor in the nuances of the crypto markets.
Researchers have observed and documented, over several decades, a few stylized facts about traditional financial markets. The propensity for markets to suddenly crash is much higher than the probability of an upward movement of similar magnitude. When markets crash the prices of most assets move in tandem or they fall together. That is, during market crashes assets tend to have higher correlations. This correlated movement of prices is due to the extensive linkages that have developed between financial markets over the years (e.g. globalization of financial markets, introduction of synthetic derivatives and options, deregulation, etc.).
Volatilities tend to be higher during market crashes. We need to take note of this point about asset prices moving a lot more, and tending to become more volatile when markets crash, as we build our weight calculation engine.
Asset weight calculations are generally driven by many inputs, but the most essential ingredients are volatilities and variances (also covariances) of assets. We adjust the volatilities and the variances (including covariances) of assets depending on how fast they are likely to change during market crashes. Hence, we term this methodology the “Velocity of Volatility and Variance” (or VVV) crash protection mechanism.
Our approach is ideally suited for crypto assets which, even during normal times, are very volatile and are also heavily correlated. Hence, we can expect much higher volatilities and correlations in the crypto investment landscape during a downturn. Our protection scheme is tailor-made for beating benchmarks when markets head downward. As we will demonstrate, our methodology also performs better than other weighting schemes over an entire bull-and-bear market cycle with no significant underperformance during upward market trends.
Returns are obtained as a direct result of bearing risk. Hence, an approach that allocates equal risk across all assets will yield a more robust weighting scheme. This approach is known as the risk parity approach. Here, the weights allocated to an asset are proportional to their corresponding risk (as opposed to their expected return) relative to that of the overall portfolio. And, this weighting mechanism is statistically measured using volatility (i.e. price movement over time) and also accounts for correlations between assets in a portfolio.
Implementing risk parity techniques for defi cryptoassets will require paying special attention to the nuances of how these markets operate. Crypto markets are more volatile and highly correlated (Refer to Tables [fig:Correlation-Matrix]; [fig:VVV-Weights-Comparison] shown in Section “Tables and Explanations”Below) compared to traditional finance. A weighting technique designed to outperform most benchmarks during a market crash while generating solid returns under other market conditions should be the recommended approach. And, VVV is our recommended approach.
The asset weights are the primary constituents required to calculate asset capacities, which determine how our rebalancing methodology would work. The second article has a discussion regarding our rebalancing methodology [LINK]. The VVV Weight calculation algorithms were among the earliest, if not the earliest, components we built and tested using historical data. Till now, the weight calculation engine could be invoked and utilized on an on-demand basis. The next set of enhancements are to be able to connect it to data updates, and completely automate them, so that these calculations can run on a daily basis or even several times during a 24-hour period.
The VVV weighting methodology factors in several empirical observations about financial markets and tailors them to the more volatile and correlated defi environment. This approach does well under a wide variety of market conditions and is custom built to outperform benchmarks during market downturns. Building portfolios in this manner epitomizes our belief that upward movements will take care of themselves, but it is the downward movements that require the most preparation. As we have discussed in our earlier article [LINK], volatility is caused by the actions of traders. It is inevitable when a huge number of traders, with different perceptions of value, transfer large sums of money. Volatility is a small price to pay for the convenience of trading anything from anywhere and anytime, as long as we are sufficiently equipped to deal with downward movements. VVV is the protection mechanism sorely needed for the defi space.
Tables and Explanations
Each of the tables in this section are referenced in the main body of the article. Below, we provide supplementary descriptions for each table. The full data sample consists of daily observations over the last 365 days. The names of the assets have been anonymized for various portfolio management considerations, though all the information presented is based on actual historical numbers.
Table ([fig:Correlation-Matrix]) — Each cell represents the correlation between the asset returns in the corresponding row and column over the historical period.
Table ([fig:VVV-Weights-Comparison]) — The total of six columns (to the right of the column, AssetName) represent the following information respectively:
- Volatility (annualized) of the assets calculated on a 90 day moving internal;
- vvvFactor calculated on a 90-day moving interval (i.e. volatility of volatility);
- VVV-Adj-Volatility, which is the sum of annualized Volatility and vvvFactor;
- vvvWeight calculated using VVV-Adj-Volatility;
- mvoWeight calculated using mean variance optimization (MVO) by Markowitz;
- noShortWeight calculated using MVO with no shorts (or no negative weight).
The last three rows show the portfolio expected return calculated based on the annualized average return over the data set horizon; the portfolio volatility; the sharpe ratio given by the portfolio expected return minus benchmark rate of 10% divided by the portfolio volatility.
Key Observation: If you determine weights using Markowitz’s MVO, you will see negative weights for certain assets. That would mean, the portfolio will need to have short positions on those assets and those derivatives might be unavailable in Defi. So, that can force the portfolio designers to eliminate those assets and compensate for the difference across all remaining assets. Net result will be such that the entire portfolio will be primarily comprised of some of the largest reserve cryptoassets (e.g. BTC, ETH, BNB, etc.). This defeats the entire premise of building a diversified portfolio in the first place. It makes very little sense to rely on the traditional weighing mechanism. In fact, we see this problem recurring in crypto such as IndexCoop’s DPI.
Conclusion of the Analysis: The portfolio based on VVV significantly outperforms the portfolios using MVO (mean variance optimization) and MVO without shorts by 74–81%. Yes, VVV Portfolio takes more risk (by 34–36%). However, for the unit of risk assumed, the portfolio based on VVV will generate a much higher return — as demonstrated by the higher Sharpe Ratio. This is one trade-off worth the risk.