Olsen
The Future Of Digital Markets
Richard Olsen
Olsen
Outline
• Distributed ledger technology• One global Internet market• Price-spread-time matching engine• Complete automation of decisions• Event-based intrinsic time• WikiFinClimate
Olsen
Today
• Financial system architecture is a pile of spaghetti....• Historically grown - step by step computerization of
manual business processes. • Delivery and settlement of trades is batch based at t+2
days from time of trade.• Every bank relies on its own book keeping. Institutions
are islands – verification of trades is cumbersome.• High uncertainty, multiplication of risk, big transaction
costs and lack of liquidity and transparency.
Olsen
Bank of England Report Q3 2014:
Distribututed Ledger Technology (DLT) is biggest innovation since discovery of double bookkeeping.
Proof of concept:Bitcoin (2009) Variations (Litecoin, Ethereum, Ripple, etc).
Distributed Ledger Technology
Olsen
……………….
Distributed Ledger Technology
A
Internet broadcast
B
Bookkeeper
Bookkeeper
Bookkeeper
Bookkeeper
Bookkeeper
Validation
The ledger
The ledger
The ledger
The ledger
The ledger
……………….10 minute cylces
Trad
e
Olsen
Benefits of DLT
• DLT for any financial asset• Distributed bookkeeping• N eyes are better than 2 eyes• No central authority• Within 10 minutes validated transaction ledger• No double spending• The ledger is the ledger is the ledger is the ledger...• Transparency, certainty and clarity
Olsen
DLT For All Assets And Issuers
Distributed Ledger Technology (DLT) with
colored coins
Transactions
Olsen
Internet Exchange
Distributed Ledger Technology (DLT)
Internet Exchange:matching system for orders,
revenue x % of average spread
Hourly Fixings WikiFinClimate
Olsen
Matching Engine
Objective: efficient price discovery• Minimizing transaction costs• Consolidation of liquidity• Orderly queueing by all market participants• Optimizing information flow
Olsen
80% Flow With <20 Min Duration
• Short-term market dynamics dictate price action.• Spread valuable indicator of private information• Expectation and price discovery
Olsen
Price-Spread-Time Queueing
Spread is indicator for private information and powerful tool to shape expectations.
Matching engines need to reward market participants for revealing private information.
See Golub, A., Dupuis, A., Olsen R. B., "High Frequency Trading Strategies in FX Markets", In High-Frequency Trading - New Realities for Traders, Markets and Regulators, Edited by Easley, D., Lopez de Prado, M., O'Hara, M., 2013, (Riskbooks)
Olsen
Effects Of Price-Spread-Time Queue
• Not same as minimum lifetime of quotes• Continuous reshuffling of queue, more stochastic• Changes speed race• Crossing of spread with one-sided prices• Low micro volatility and efficient price discovery
Olsen
Benefits Of Internet Exchange
• 10 minute settlement• Low transaction costs • Any ticket size• All asset classes and issuers• All cross rates available• Emergent money• Liquidity• Efficient, fair, transparent
Olsen
Complete Automation Of Decisions
• Today, models need handholding.• Trading glitches (Flash crash, Knight Capital, etc.)• Future full automation• Shift of academic work to real time testing• We need to come up with powerful models!
Olsen
Necessity To Rethink Time
Olsen
Relativity Theoy: Twin Paradox
Is economics is a multi-system problem?
Olsen
How To Sample A Time Series?
• Tick-by-tick?• Every second?• Every minute, hour, day, week, month…data?• How to interpolate, if no data is available?
Olsen
Physical Time Is Static
Mapping of data with physical time grid loses information about extremes.
Olsen
Frequency Of Sampling
Increasing Frequency Of Observations Adds Noise
Olsen
Increased Sampling Reduces Signal Quality
• Basic problem: information is in tails.• Signal to noise ratio deteriorates with increased sampling.
Signal
Noise
Olsen
Time Issues• Sampling and testing in physical time • Uniqueness of events• Sampling frequency and length of coastline• Consistent time aggregation across time scales • Impact of seasonality and heatwave effects
Model quality is as good as definition of time.
10 reaserch papers or so discuss time...low citation number.
Olsen
Physical Time
• Physical time maps the rotation of the earth. • It is a uniform scale: X = (X_1, X_2, X_3 …. )• Events have equal weights.• There are fixed equidistant time intervals of 1 minutes, 1
hour, 1 day, 1 week.
Olsen
Event Time: Reversal From Extreme
An event is defined as a price reversal from extreme by x %.
In our papers we call a price reversal a directional change.
Olsen
Intrinsic Event Time
• Uniform intrinsic time scale for a given reversal of x percent: X = (X_1, X_2, X_3 …. )
• Events have equal weights in intrinsic time• Events occur after reversal of x percent from extreme.• Intrinsic time scales of different thresholds are analogous
to 1 minute, 1 hour, 1 day, etc. time scale.
Olsen
Empirical Evidence
Overshoots are on average equal to threshold; this is true for all observed thresholds: scaling law.
Valuable ex ante information
Olsen
Schemata Event Time
Olsen
Physical versus Event Time
Physical timee.g. [hours]
1 2 3 4 5 6
Intrinsic time[events]
1 2 3 4
Dx
-Dx
-Dx
Dx
Olsen
Static Versus Dynamic View
Physical timee.g. [hours]
1 2 3 4 5 6
Intrinsic time[events]
1 2 3 4
Dx
-Dx
-Dx
Dx
Assymetric uncertainty
Olsen
Established scaling laws
Müller et al., J. Bank Finance, 1990: Mean absolute change of mid-price to time
Guillaume et al., Finance Stoch. 1997:Number of directional changes to thresholds
Olsen
Scaling laws
Decomposing total price move into directional-change and overshoot:
Leads to 9 additional scaling laws:
Olsen
Tick-count scaling law
Olsen
Benefits Of Scaling Laws
• Short data samples for estimation• Dynamic frame of reference• Grid of scaling laws – combine information • Event language as function of overshoot
Olsen
Extrema
Exposure
1% 1%
Observe Manage
Coastline TradeTake Profit
Take Profit
Average
Open
Average Overshoot
Agent-based Models
Olsen
Agent Models With Scaling Laws
• Dynamic investment strategies• Liquidity provision• Price stabilizing • Outlook: 85 percent dynamic strategies
Olsen
WikiFinClimate
• Wikipedia for keywords: 4.6 Mio articles in English • WikiFinClimate for time series• Complete Automation Of Decisions• Crowd-based modeling• Real time information service• Graphical visualization, etc.
Olsen
WikiFinClimate
Distributed Ledger Technology
Time Series
Internet Exchange
Crowd-based information system
Olsen
WikiFinClimate
Contributors
modeling, simulation, real time operation
Model A Model … …………… Lego block 4 ……………..
Users
Olsen
Conclusion
• One global Internet market with DLT• Price-spread-time matching engine• Complete Automation Of Decisions• Event-based intrinsic time• Agent-based models with scaling laws•WikiFinClimate• Lots of open questions, work....
Top Related