Acaduneia:HowAcademicsareUsingDune

Dune Wizards are a diverse group. They come from all backgrounds, geographies, and skill levels - and do their magic everywhere from spare bedrooms to the halls of institutional power.

Some are first-time analysts, taking their first steps into data through Dune. Others have vast data science experience and use Dune as their go-to tool for checking the chain.

Others are some of the top academic & research institutions in the world!

Over the past few months, we’ve seen several interesting papers come out of the academic and think tank sphere, from well-known institutions like:

1. Stanford
2. Cornell
3. UPenn
4. LBS
5. The IMF
6. Bank of Canada
7. Scalability and fragmentation in crypto payment systems.
8. NFT product/market fit and pricing theories.
9. Crypto standards development.
10. Slippage reduction in AMM trading.
11. The (inevitable?) centralization of governance tokens.
12. Frederic Boissay, Giulio Cornelli, Sebastian Doerr, and Jon Frost
13. ‍Blockchain scalability and the fragmentation of crypto
14. ‍Bank for International Settlements Bulletin No 56
15. Blockchains rely on native token prices to ensure security.
16. High token prices drive new entrants to alternative chains, leading to fragmentation.
17. Thus, as a blockchain grows it experiences anti-network effects (more users = pressure to shrink the network)
18. Money is a social coordination tool, and traditional payment methods and networks grow through network effects - thus “there is more promise in innovations that build on trust in sovereign currencies.”
19. How can we better understand users' flow from one chain to another based on chain usage and native token price?
20. What metrics do we have and need to measure the monetary qualities of crypto assets?
21. What does the industry need to have better interoperability? What data can we surface to support those efforts?
22. Roman Kräussl, Universite du Luxembourg - Department of Finance; Hoover Institution, Stanford University
23. Alessandro Tugnetti, University of Luxembourg - Department of Finance
24. ‍Non-Fungible Tokens (NFTs): A Review of Pricing Determinants, Applications and Opportunities
25. ‍SSRN
26. NFTs share the investment profile of physical collectibles, namely a high yield with a high-risk profile.
27. NFT prices move with BTC and ETH, in part because many people move their crypto wealth into NFTs - to speculate on an investment, to express their passion for a particular creator or collection, and to become members of the virtual communities associated with certain NFTs.
28. Because NFT markets are even newer than cryptocurrencies, they’re even earlier on the price discovery curve and thus very inefficient markets; we don’t have a definitive NFT pricing model as we do for fine art.
29. The oracle problem: there have been relatively few attempts to link NFT pricing to real-world data like real-time currency exchange rates to better price NFTs, or to check the delivery status of a physical product that comes with an NFT purchase to enable an extra level of buyer protection.
30. The interconnection problem: though token bridges have somewhat solved the cross-chain transfer challenge for fungible tokens, the lack of cross-chain transferability for NFTs means the markets on different chains have to, at least in part, be considered separately when thinking about pricing.
31. What data and metrics can we use to understand and define the key properties of each main category of NFTs?
32. How about the pricing models discussed? What dashboards can be created to compare different collections to themselves or others through the lenses of hedonic regression models, repeat sales regressions, vector autoregressive models, machine learning, and wavelet models?
33. How can we combine Dune blockchain data with other sources to tell a complete picture (e.g. the Google Trends to Unique NFT Wallets graph)?
34. Sarah Hammer, The Wharton School of the University of Pennsylvania
35. Brett Hemenway Falk, University of Pennsylvania - Department of Computer and Information Science
36. ‍Taming the Wild West: Achieving Public Policy Goals through Crypto Standards
37. ‍SSRN
38. They can reduce development costs by allowing multiple organizations to reuse (standardized) smart contract code.
39. They enable interoperability (see the 60k+ tokens that can be swapped on Uniswap thanks to ERC-20)
40. They can be made global, similar to the Unicode character encoding standards, to facilitate international trade and cooperation.
41. Since they’re voluntary, industry-wide input is needed for their development and formalization so the industry itself has a vested interest in compliance.
42. Since they don’t have the force of law, the potential repercussions of parties not adhering to standards have to be considered.
43. Standards implemented on-chain cannot verify the content of the data put on-chain via oracles - so issues around accuracy and fraud will still exist.
44. On-chain standards can’t control off-chain activities, so compliance and enforcement will have to be handled somehow.
45. How widely used are OpenZeppelin’s “reference implementations?” Can their adoption rate be analyzed to examine the potential opportunities and challenges in creating blockchain standards?
46. More and more centralized exchanges are putting proof of reserves on-chain - what metrics need to be monitored to contextualize this information and provide early warnings if a CEX may be in trouble?
47. What are the key oracles relied on by the crypto industry? How can we see the extent to which certain oracles are referenced?
48. ‍Irene Aldridge, Cornell University
49. ‍Slippage in AMM Markets
50. ‍SSRN
51. Positivity in the quantities of both instruments to ensure that positive quantities are traded.
52. Convexity, to ensure that the prices rise when liquidity falls.
53. Asymptotic properties with respect to both x and y axes to ensure support for potentially infinite liquidity.
54. A review of the key differences between Automated Market Making (AMM), which is the most popular format for Decentralized Exchanges, and traditional limit-order book-based market making.
55. A step-by-step analysis of crypto-token price formation.
56. A proposal of one solution for ex-ante slippage estimation that can be used to better trade on AMMs.
57. Can you identify trades with positive and negative slippage and compare by tokens traded, DEXs, and time?
58. How do trading volumes relate to Liquidity Pool sizes? Do some pools have much more liquidity than they need to minimize slippage given their average volume?
59. What can we infer about Orderbook vs AMM trade models based on the usage of each in different protocols?
60. Joaquin Delgado Fernandez, University of Luxembourg
61. Tom Josua Barbereau, University of Luxembourg
62. Orestis Papageorgiou, University of Luxembourg
63. ‍Agent-based Model of Initial Token Allocations
64. ‍University of Luxembourg
65. Does trading behavior affect voting rights token distributions over time?
66. Do alternative, ’fair launch’ token allocations affect voting rights token distributions over time?
67. An agent-based model for the analysis of token distributions under various market conditions reflective of trading.
68. Simulation results showing how over time, regardless of initial token allocation, concentration is imminent.
69. Extended understandings of tokenomics to formerly include token allocations as part of governance parameters.

Though they conclude the fair launch allocation pioneered with $YFI did not effectively combat governance token concentration, they acknowledge a few limitations:

1. Their clearing mechanism lacks a formal price clearing method, and they didn’t implement limit orders.
2. In their model, the decision-making of an individual agent does not depend on past decisions or those of other agents.
3. They rely heavily on the Fear Greed Index as a proxy for market conditions instead of more granular indicators like specific asset prices or social media data.
4. How do the different distribution methods outlined here compare to each other in the real world? (Schedule-based; Pre-mined, scheduled distribution; Pre-mined, one-off distribution; Discretionary)
5. What about the allocation or incentive systems? (See Table 7.)
6. How do Voting Escrow (“ve”) tokens like veCRV, which are less liquid tokens used for voting, compare to non-escrowed, highly liquid governance tokens?

What an interesting research area! This is just a small sample of more than 20 research papers that have come across the Dune desk in recent months. There's so much interesting work being done, and so much more for all of us to explore too.

The great thing about blockchain data though, is that you don’t need to be a PHd candidate at a prestigious institution to access and crunch it with professional tools. All you need is some SQL chops and a Dune account.

Anon Wizards can put out analysis just as good as Ivy League Professors!

It’s great to see these institutions really diving deep into the frontiers of finance, and we’re looking forward to more papers.

Make no mistake about it - what’s happening here on the frontiers of finance is very interesting to those who walk the halls of higher institutions - but learning, understanding and analyzing it isn’t limited to those halls anymore........