One of such strategies is to match actual on-chain transaction metrics with off-chain investment activity.
First, we note that ETH and BTC are the currencies that handled most of the Ripple network operations (transactions and trading) during November and December (each one leading or lagging respect to the other at different moments).
nov
In our predictive model we use transaction count as a measure of network activity (what we want to predict), and, two services used in common by the users of the most popular XRP wallet during both months (X1 and X2, our drivers). X1 is an instant exchange (of the real-time centralized type), and, X2 is a prominent Ripple gateway. The resulting formulas are as follows,
In this model, the usage of the instant exchange has the higher positive impact — this may indicate that November was partly a period of recreational investment, no large individual investment activity would have taken place in instant exchanges (due to more competitive fees, this would likely have happened in a traditional exchange or in a gateway). There are also several cyclical components in the formula (sin, cos) that suggest having some degree of seasonality/business-as-usual patterns.
dec
In this case, gateway trading has a positive impact on price, instant exchange trading a negative impact. This is consistent with the substantially higher volumes during December.
Time series analysis and prediction
Temporal correlations in blockchain traffic
volatility of volatility (fragility test)
fragile: has to be non-linear to harm (has to be accelerated to harm)
second derivative of "trust" distance metric (visits per day, the event size)
or model directly the second derivative with nutonian?
Information flow
Information flow and information leakage in blockchain services for resources allocation (Golem, etc.)
Law's expected force
Optimization of blockchain configurations based on social contexts groups (social-prediction markets, liquidity providers, etc.)
social fabric fills the role of communication: you can improve by introspection if you communicate
Network effect in on-chain\off- chain interplay
model active accounts driven by social network activity
Spatio-temporal patterns in blockchain networks
Vector fields (temporal streams)
it must be decoded sequentially over time
value by memory
As a particular type of language, the "static" (neglecting random transcription errors, recombination and mutation) DNA and its transcription pattern over time yields biologically essential s.t. patterns.
e.g. computer languages, are not read and interpreted in one step, but sequentially, thus, their meaningfully arranged vocabulary (e.g. "computer code") can be seen as a s.t. pattern.
Production of space
blockchains are production spaces (Forging Blockchains: Spatial Production and
Political Economy of Decentralized
Cryptocurrency Code/Spaces
Joe Blankenship)
exploitation of automated robot labor
developers as the dominant class within the social and technical spaces of the blockchain technology
have ultimately leveraged their knowledge/power dynamics to accumulate wealth via the token value, and
shifted them into the role of investor.
communication and social fabric (abstract, tacit)
automation and obfuscation of themechanisms (of production)
geographic borders via the conflicting abstract conditions (social,political, and economic) w
qualitative context of the social dynamics
Circles of trust
People trust in structure
you generally do not care about who wrote the health article as long as "structure" suggest he is not a charlatan
theory of the firm reimagined: you do business in your circle where trust is secured
Providing security and privacy for social overlay networks in blockchain protocols
Configuring blockchain protocols' parameters based on the networks' topology analysis
learn the distance, learn the trustability
Use of low complexity property testing methods by decentralized blockchain agents
Neighborhood Asymmetry
function which sums the vectors from the dataRecord to the neighbors implictly defined by the supplied dataMatrix and returns the length of this resulting neighborhood directionality normalized by the number of neighbors. Thus, this metric is primarily concerned with the symmetry of the neighbor distribution but also contains a contribution from the distance to each of the neighbors. In this form, a 2-norm is used to compute the distances.
InformationMetric ... measure the information content of each sample.
measure the maximum distance between any one of the variables
response variable is fees