Produced by The TIE

 



1 Introduction

The cryptoasset market is different than traditional financial markets, in which important variables tend to operate under persistent relationships that are well-studied and generally understood. Crypto, on the other hand, is highly dynamic and operates in an innovative environment. Industry shifting developments occur at a high frequency and market analyses must be conducted on an ongoing basis in order to remain current. In order stay on top of the market at large and monitor the evolving relationships among cryptoassets, The TIE collects a number of financial and social media measures to summarize current market conditions in both financial and social media domains.

These measures include, but are not limited to:

Financial:

  • Returns: Assess sector performance over the week
  • Volatility: Assess volatility across sectors
  • Correlations: Monitor shorter-term relationships between returns
  • Cointegrations: Monitor longer-term relationships between prices
  • Trading volumes: Assess trading activity
  • Returns by geography: Assess market activity in different areas of the world

Social media:

  • Twitter firehose: Monitor sector and asset chatter over the week
  • Twitter influencers: Monitor influencer conversation over the week
  • YouTube influencers Monitor influencer content over the week


1.1 Sector Allocations


Many outputs will be assessed on the sector level. As such, sector portfolios were created to assess sector trends and quantify convergence/divergence between important categorizations of cryptoassets.

Methodology: Each sector portfolio contains a market-cap-weighted basket of up to eight assets that individually comprise no more than 1% of its sector.

Below are the categorizations for data through 2021-11-07.


Allocations by sector:


Currencies
DEXs/AMMs
Gaming/Metaverse
Layer 1s
Oracles/Storage/Web3
Privacy
Bitcoin 91% Uniswap 42% Axieinfinity 48% Ethereum 70% Chainlink 34% Monero 37%
Xrp 4% Elrond 15% Flowdapperlabs 26% Cardano 10% Vechain 21% Zcash 16%
Dogecoin 3% Pancakeswap 12% Enjincoin 9% Solana 9% Filecoin 18% Dash 15%
Litecoin 1% Aave 11% Decentraland 9% Polkadot 6% Iota 9% Decred 12%
Bitcoincash 1% Thorchain 7% Thesandbox 4% Avalanche 2% Bittorrent 6% Horizen 7%
Compound 5% Wax 3% Algorand 2% Arweave 4% Nucypher 5%
Curve 4% Tezos 1% Holochain 4% Digibyte 5%
Sushi 4% Nearprotocol 1% Siacoin 2% Verge 3%



2 Market Snapshot


Did DEXs outperform Web3 last week? How did 15-day volatility vary among sectors last week? Did correlations with BTC/ETH go up or down last week?


2.2 Sector Performance


Methodology: Volume weighted average price computed using The TIE methodology.

The below plot depicts sector cumulative returns (%) for the seven days prior to 2021-11-07.



2.3 Sector Volatility


Volatility measures the degree to which price moves. Assets with prices that fluctuate wildly have high measures of volatility.

Methodology: Volatility is the standard deviation of log returns for the market-cap-weighted sector portfolios.



2.4 Sector Correlation


Correlation measures the degree to which two normally distributed variables move in relation to each other. It is an associational statistic that falls between -1 and 1. Intuitively, correlation between cryptoassets is a short-term relationship between returns.

Methodology: Correlation refers to the correlation of log returns for the market-cap-weighted sector portfolios.


2.4.1 With Bitcoin


2.4.2 With Ethereum





4 Market deep dive


4.1 Network Visualization

In order stay on top of the changing relationships between cryptoassets, The TIE conducts additional correlational analysis to monitor groups of cryptoassets that are currently behaving as a community. Understanding the relationships between cryptoassets from a quantitative perspective can not only lend additional credence to qualitative understandings of market dynamics, but also reveal when those qualitative assumptions are not reflected in the data.

Cryptoassets grouped in the below plot indicate similar return and performance characteristics over the past few months.

Methodology: Because correlation estimates between log-return series can be misleading (spurious correlation, confounding variables), The TIE conducts a regularization technique that is based on the eigen decomposition of the sample correlation matrix. In order to reduce the influence of noise in the data, we regularize the sample correlation matrix by approximating it by a low-rank matrix based upon the first 10 eigenvectors. We then consider the statistics as an undirected network of associations between cryptoassets, and apply community detection algorithms to quantitatively ascertain clusters in the data.


4.2 Bitcoin, Ethereum, Sector returns by Region

The below plot depicts cumulative returns for Bitcoin, Ethereum, and top-level sectors for common trading hours in different areas of the world.


4.3 Sector Cointegration

Cointegration tests measure the degree to which the price series of two cryptoassets move together. If two cryptoassets are cointegrated, then the difference between their prices tends to be similar over time. In other words, the price series don’t move in different directions for long before reverting to their typical spread.

Whereas correlation is a short-term relationship between returns, cointegration is a long-term relationship between the prices. Cointegration can be seen as a measure of similarity of assets in terms of risk exposure profiles.

Methodology: Johansen procedures were conducted on the price series for Bitcoin, Ethereum, and the market-cap-weighted sector portfolios.

Interpretation: The blue line depicts a test statistic for a measure of cointegration. The black horizonal line depicts the critical value of the test statistic at a significance level of 0.05. If the blue line is above the black line, we say that the price series’ are cointegrated for the previous 30 days.


4.3.1 Bitcoin and Ethereum

4.3.2 Bitcoin and Sectors

4.3.3 Ethereum and Sectors


5 Appendix

Note: This section contains additional outputs that are not presented in the body of the report due to their granularity or focus on a specific section in the market.

5.1 Sector Overlay Correlation and Cointegration


5.1.1 Bitcoin and Ethereum

5.1.2 Bitcoin and DEXs/AMMs

5.1.3 Bitcoin and Gaming/Metaverse

5.1.4 Bitcoin and Layer 1s

5.1.5 Bitcoin and Oracles/Storage/Web3

5.1.6 Bitcoin and Privacy

5.1.7 Ethereum and DEXs/AMMs

5.1.8 Ethereum and Gaming/Metaverse

5.1.9 Ethereum and Oracles/Storage/Web3

5.1.10 Ethereum and Privacy