Content uploaded by Ignacio Ozcariz
Author content
All content in this area was uploaded by Ignacio Ozcariz on Apr 30, 2020
Content may be subject to copyright.
QUANTUM SUPREMACY :
in
FINANCIAL MARKETS (FOREX)∗
Ignacio Ozcariz†
RQuantech - Geneva (Switzterland)
Criptosasun - Madrid (Spain)
‡
(Dated: November 7, 2019)
The paper presents how to achieve market dominance, implementing the oldest plain vanilla price
arbitrage strategy, using the newest quantum entanglement phenomena.
CONTENTS
I. Introduction. 1
II. Dynamics of FOREX (FX) Market 1
A. High Frequency Trading (HFT) 1
B. Direct Market Access (DMA) 2
C. Strategies 2
1. Latency-Arbitrage 2
2. Regulation (USA) 2
D. Summary 2
III. Quantum Information and Quantum Computation3
A. Introduction (Ref:[1]) 3
B. Quantum State – Vector representation
(Ref:[2]) 3
C. Measurements (Ref:[2]) 3
D. Entanglement (Ref:[2]) 4
IV. Methodology 4
A. Set-up 4
B. Price Communication 4
C. Price Reception and Comparison with current
prices 4
D. Quantum Processes 5
E. Summary 7
V. Implementation 7
A. Structure and Technologies 7
1. Tick-to-Trade 8
2. Communications 8
3. Quantum supremacy 8
B. State of the Art 8
C. Disclaimer 8
VI. Conclusions 8
Acknowledgments 8
References 8
∗Applicable also to any financial market
†Also at Universidad Politecnica Madrid. Doctorate Student;
im.ozcariz@alumnos.upm.es
‡i.ozcariz@rquantech.com
I. INTRODUCTION.
“It from bit” symbolizes the idea that every item
of the physical world has at bottom—a very deep
bottom, in most instances — an immaterial source and
explanation; that which we call reality arises in the last
analysis from the posing of yes-or-no questions and the
registering of equipment-evoked responses; in short, that
all things physical are information-theoretic in origin
and that this is a participatory universe.”
– John A. Wheeler, [3]
Thirty years later from this statement it’s time to up-
date it and propose that “It from qubit” could be the
last truth.
This paper will present a novel approach to the
use of quantum techniques that we nickname quantum
supremacy applied to the financial markets and particu-
larly to the FOREX market.
We’ll demonstrate that having two entangled devices,
more on the definition of entanglement in Section 3 of
the paper, it’s possible to achieve a full dominance in the
FOREX using a very simple plain vanilla strategy, price
arbitrage between two far away markets.
II. DYNAMICS OF FOREX (FX) MARKET
Each day, billions of monetary units are exchanged on
the foreign exchange currency market. FX trading is used
to determine currency exchange rates across the world.
While people have been trading currencies for thousands
of years, modern technology has changed the way that
many banks and individual investors do business.
One of the key technologies in the evolution of the
markets has been high frequency trading (HFT) and its
impact in the FX trading has been extraordinary.
A. High Frequency Trading (HFT)
According to the dictionary of financial terms of NAS-
DAQ [4]
High Frequency Trading (HFT)
2
Refers to computerized trading using proprietary algo-
rithms. There are two types high frequency trading.
Execution trading is when an order (often a large or-
der) is executed via a computerized algorithm. The pro-
gram is designed to get the best possible price. It may
split the order into smaller pieces and execute at different
times.
The second type of high frequency trading is not exe-
cuting a set order but looking for small trading opportu-
nities in the market.
It is estimated that 50 percent of stock trading volume
in the U.S. is currently being driven by computer-backed
high frequency trading. Also known as algo or algorith-
mic trading.
B. Direct Market Access (DMA)
Access to the markets was historically being mediated
by banks or direct brokers at the markets. From the last
decennials of the past century this interrelation has given
pass to the direct access to the markets named DMA
(Direct Market Access).
According to David Polen [5] :
There are two distinct market segments that use DMA
- the human trader and the blackbox. I like to call
this “Human DMA” and “Highfrequency Trading (HFT)
DMA”.
With Human DMA, the extreme is a buy-side that has
traders manually executing trades and looking at market
data over the Internet-
With HFT DMA, the extreme is a blackbox co-located
at the exchange. One market segment is sub-millisecond
and the other is more than tens of milliseconds - some-
times hundreds of milliseconds.
C. Strategies
About the strategies used by the black-boxes, in 2014
the Security and Exchange Commission (SEC) appears to
emphatically say that two strategies, order anticipation
and momentum ignition, are manipulative and illegal.
The SEC writes [6]:
“Directional strategies generally involve establishing a
long or short position in anticipation of a price move
up or down. The Concept Release requested comment
on two types of directional strategies – order anticipa-
tion and momentum ignition – that ‘may pose particular
problems for long-term investors’ and ‘may present seri-
ous problems in today’s market structure.’
An order anticipation strategy seeks to ascertain the
existence of large buyers or sellers in the marketplace and
then trade ahead of those buyers or sellers in anticipation
that their large orders will move market prices (up for
large buyers and down for large sellers).
A momentum ignition strategy involves initiating a se-
ries of orders and trades in an attempt to ignite a rapid
price move up or down.
As noted in the Concept Release[7], any market partic-
ipant that manipulates the market has engaged in con-
duct that already is illegal. The Concept Release focused
on the issue of whether additional regulatory tools were
needed to address illegal practices, as well as any other
practices associated with momentum ignition strategies.”
1. Latency-Arbitrage
Going down to the technicalities, Latency Arbitrage is
an important concept when discussing High Frequency
Trading, and refers to the fact that different people and
firms receive market data at different times.
These time differences, known as latencies, may be as
small as a billionth of second (a nanosecond), but in the
world of high speed trading, such differences can be cru-
cial. So crucial, in fact, that trading firms pay lots of
money to be located closer to exchanges’ servers– each
foot closer saves one nanosecond.
Latency arbitrage occurs when high frequency trading
algorithms make trades a split second before a competing
trader, and then resell the stock seconds later for a small
profit.
2. Regulation (USA)
The previous Congress dealt with potential regulation
of HFT (US Congress HFT Overview of Recent Devel-
opments 2016)[8]. We very briefly refer here to the Tes-
timony on Regulatory Reforms to Improve Equity Mar-
ket Structure by Stephen Luparello, director, Division
of Trading and Markets, U.S. Securities and Exchange
Commission, Before the Senate Committee on Banking,
Housing and Urban Affairs Subcommittee on Securities,
Insurance and Investment. This declaration was made
March 3, 2016.
Luparello said SEC staff was developing a recommen-
dation for the SEC to consider addressing the use of ag-
gressive, destabilizing trading strategies that could exac-
erbate price volatility.
Until today regulation is still a working process.
D. Summary
It can be extremely challenging to earn a profit when
trading currency. In many cases, investors can lose sig-
nificant sums of money. Exchange rates can be impacted
by a variety of factors. This can include economic condi-
tions, politics, weather, shipping conditions, piracy, tech-
nology advances and more.
Many HFT programs are installed in specialized data
centres located near an exchange (Co-location). Since the
speed of execution is limited by the speed of light, many
programmers and investors try to minimize the amount
3
of time it takes for an order to be executed. This is
possible by minimizing the amount of time it takes data
to travel between the operator premises and an exchange.
Most HFT programs are designed to profit from very
small price differences in a currency. In many cases, a
program will make a profit of only a few cents per trade.
However, millions of these types of trades every day can
yield a significant profit.
Our approach in this paper we’ll be to use the oldest
strategy, price arbitrage between two locations, profit-
ing the quantum supremacy. (The spooky action at a
distance; in Einstein words).
III. QUANTUM INFORMATION AND
QUANTUM COMPUTATION
A. Introduction (Ref:[1])
a. Quantum computing fundamentals. All comput-
ing systems rely on a fundamental ability to store and
manipulate information. Current computers manipulate
individual bits, which store information as binary 0 and
1 states. Quantum computers leverage quantum mechan-
ical phenomena to manipulate information. To do this,
they rely on quantum bits, or qubits.
b. Quantum Properties. Three quantum mechan-
ical properties — superposition, entanglement, and
interference — are used in quantum computing to
manipulate the state of a qubit.
1. Superposition
Superposition refers to a combination of states we
would ordinarily describe independently. To make
a classical analogy, if you play two musical notes at
once, what you will hear is a superposition of the
two notes.
2. Entanglement
Entanglement is a famously counter-intuitive quan-
tum phenomenon describing behavior we never see
in the classical world. Entangled particles behave
together as a system in ways that cannot be ex-
plained using classical logic.
3. Interference
Finally, quantum states can undergo interference
due to a phenomenon known as phase. Quantum
interference can be understood similarly to wave
interference; when two waves are in phase, their
amplitudes add, and when they are out of phase,
their amplitudes cancel.
B. Quantum State – Vector representation
(Ref:[2])
What then is a qubit?
Just as a classical bit has a state – either 0 or 1 – a qubit
also has a state. Two possible states for a qubit are the
states |0iand h1|, which as you might guess correspond
to the states 0 and 1 for a classical bit.
Notation like | i and h | is called the Dirac notation,
and we’ll be seeing it often, as it’s the standard notation
for states in quantum mechanics. The difference between
bits and qubits is that a qubit can be in a state other than
|0ior h1|. It is also possible to form linear combinations
of states, often called superpositions:
|ψi=α|0i+β|1i(1)
FIG. 1. Qubit Bloch Sphere representation
The numbers αand βare complex numbers, although
for many purposes not much is lost by thinking of them
as real numbers. Put another way, the state of a qubit
is a vector in a two-dimensional complex vector space.
The special states |0iand h1|are known as computational
basis states and form an orthonormal basis for this vector
space.
We can examine a bit to determine whether it is in
the state 0 or 1. For example, computers do this all the
time when they retrieve the contents of their memory.
Rather remarkably, we cannot examine a qubit to deter-
mine its quantum state, that is, the values of αand β.
Instead, quantum mechanics tells us that we can only ac-
quire much more restricted information about the quan-
tum state. When we measure a qubit we get either the
result 0, with probability |α|2, or the result 1, with prob-
ability |β|2.Naturally, |α|2+|β|2= 1, since the probabil-
ities must sum to one. Geometrically, we can interpret
this as the condition that the qubit’s state be normalized
to length 1. Thus, in general a qubit’s state is a unit
vector in a two-dimensional complex vector space.
C. Measurements (Ref:[2])
Suppose we have two qubits. If these were two classical
bits, then there would be four possible states, 00, 01, 10,
and 11. Correspondingly, a two qubit system has four
computational basis states denoted |00i,|01i,|10i,
|11i.
4
A pair of qubits can also exist in superpositions of these
four states, so the quantum state of two qubits involves
associating a complex coefficient – sometimes called an
amplitude – with each computational basis state, such
that the state vector describing the two qubits is
|ψi=α00|00i+α01 |01i+α10|10i+α11 |11i(2)
Similar to the case for a single qubit, the measurement
result x (= 00, 01, 10 or 11) occurs with probability
|αx|2, with the state of the qubits after the measure-
ment being |xi. The condition that probabilities sum to
one is therefore expressed by the normalization condition
that Px∈{0,1}2|αx|2= 1,where the notation x∈ {0,1}2
means ‘the set of strings of length two with each let-
ter being either zero or one’. For a two-qubit system,
we could measure just a subset of the qubits, say the
first qubit, and you can probably guess how this works:
measuring the first qubit alone gives 0 with probability
|α00|2+|α01 |2, leaving the post-measurement state
|ψi=α00|00i+α01 |01i
p|α00|2+|α01 |2(3)
Note how the post-measurement state is re-normalized
by the factor p|α00|2+|α01|2so that it still satisfies the
normalization condition, just as we expect for a legiti-
mate quantum state.
D. Entanglement (Ref:[2])
An important two qubit state is the Bell state or EPR
pair, β00, that we’ll make use in our methodology section.
β00 =|00i+|11i
√2(4)
This innocuous-looking state is responsible for many sur-
prises in quantum computation and quantum informa-
tion. The Bell state has the property that upon mea-
suring the first qubit, one obtains two possible results: 0
with probability 1/2, leaving the post-measurement state
|φ´i=|00i, and 1 with probability 1/2, leaving |φ´
i=|11i.
As a result, a measurement of the second qubit al-
ways gives the same result as the measurement of the
first qubit. That is, the measurement outcomes are cor-
related.
These correlations have been the subject of intense in-
terest ever since a famous paper by Einstein, Podolsky
and Rosen, in which they first pointed out the strange
properties of states like the Bell state. EPR’s insights
were taken up and greatly improved by John Bell, who
proved an amazing result: the measurement correlations
in the Bell state are stronger than could ever exist be-
tween classical systems. These results were the first inti-
mation that quantum mechanics allows information pro-
cessing beyond what is possible in the classical world.
This “amazing” correlation will be at the base of our
Quantum Supremacy over the FOREX market.
IV. METHODOLOGY
A. Set-up
Market strategy will be made in the FOREX market
over London and New York markets based on the pair
EUR/USD. Let’s name the price of this pair P.
For the implementation of our strategy we’ll use the
Alice and Bob characters invented by Ron Rivest, Adi
Shamir, and Leonard Adleman in their 1978 paper ”A
method for obtaining digital signatures and public-key
cryptosystems. In our scenario we suppose that A and
B (also known as Alice and Bob) are two market ma-
chines located (collocated) in London (Alice) and New
York (Bob).
Initially, at the start of the market session, both ma-
chines shared several entangled pair of qubits (2N). Let’s
name for the first interchange the qubits Aq1,Aq2 for
Alice and Bq1,Bq2 for Bob.
Aq1,Bq1 are a pair of Bell’s qubits β00 and the same for
the other pair Aq2,Bq2.
Let’s establish a reference via the GPS clock shared by
both parties, using 20 Hz cycle, so each 50 milliseconds,
and name this time reference T0.
B. Price Communication
At this time A and B sends the current price Pof its
market to the other. Taking in account the 7,000 Kilo-
metres distance the time taken to the signal to arrive to
the other end is approx. (fibre-optics or radio transmis-
sion could made differences in the final time), 43 mil-
liseconds. Let’s name these prices P l0(London Price)
and P n0(New York price) at T0.
The potential outcomes that A and B will have regard-
ing the prices at T0are:
1. P l0=P n0State φ0(London equal price than New
York)
2. P l0> P n0State χ0(London greater price than
New York)
3. P l0< P n0State ψ0(London lesser price than New
York)
An arbitrage strategy could be established in the states
χand ψbut not in φ.
We’ll introduce a 5 milliseconds guard time to cover
potential delays of the signal due to any circumstances.
C. Price Reception and Comparison with current
prices
At T0+48 msec. that we’ll consider as T1, A and B had
new prices of Pat their respective markets. Let’s name
these new prices P l1(London) and P n1(New York).
5
FIG. 2. Initial Price States
It’s obvious that if the price in London is greater than
in NY (state χ0) a very simple strategy consists to buy
the pair in NY and to sell it in London that will carry on
a profit in the transaction. The same occurs if London
is lesser than in NY (state ψ0). In his case the strategy
consists to sell the pair in NY and to buy it in London
and the strategy will carry also on a profit.
To be successful in this strategy we need:
1. The prices when both parties received the infor-
mation regarding the other partner side have not
changed to the opposite condition. So for example,
at time T1,P l1> P n1
2. Both parties can communicate the other side the
condition 1 satisfied in its leg of the market.
The second condition allows that both parties go forward
in the buy/sell process doing the transactions concur-
rently.
Logically a further “classical” communication regard-
ing condition 2 will introduce a new delay of 43 millisec-
onds allowing the markets to change its positions during
the delay and removing any potential benefit.
At this point we’ll introduce the Quantum Supremacy
mechanism based on the qubits shared previously be-
tween A and B.
D. Quantum Processes
At the moment T1we’ll have, related to the potential
movements of the market, the next conditions:
Alice side:
If State χ0
1. Condition GO:P l1≥P l0. The market is in the
same or more favourable conditions that in T0
2. Condition NO-GO:P l1< P l0. The market is in
less favourable conditions that in T0
If State ψ0
1. Condition GO:P l1≤P l0. The market is in the
same or more favourable conditions that in T0
2. Condition NO-GO:P l1> P l0. The market is in
less favourable conditions that in T0
FIG. 3. Alice at T1
Bob side:
If State χ0
1. Condition GO:P n1≤P n0. The market is in the
same or more favourable conditions that in T0
2. Condition NO-GO:P n1> P n0. The market is in
less favourable conditions that in T0
6
If State ψ0
1. Condition GO:P n1≥P n0. The market is in the
same or more favourable conditions that in T0
2. Condition NO-GO:P n1< P n0. The market is in
less favourable conditions that in T0
FIG. 4. Bob at T1
The manipulations of the qubits that Alice and Bob will
make with their qubits depending of the previous condi-
tions are:
Alice makes a measure in Aq1that is:
1. State φ0– Do nothing
2. States χ0and ψ0– and Condition GO Vmakes
a measure using M=|1ih1|
3. States χ0and ψ0– and Condition NO-GO V
makes a measure using M=|0ih0|
FIG. 5. Alice Measurement
Bob makes a measure in Bq2that is:
1. State φ0– Do nothing
2. States χ0and ψ0– and Condition GO Vmakes
a measure using M=|1ih1|
3. States χ0and ψ0– and Condition NO-GO V
makes a measure using M=|0ih0|
FIG. 6. Bob Measurement
7
Taking in consideration the characteristics of the measure
processes described in the Section (3), after the measure
of Alice and Bob of their qubits Aq1and Bq2with the
Matrix |1ih1|the qubits will finish in the state |1iand
after the measure of Alice and Bob of their qubits with
the Matrix |0ih0|the qubit will finish in the state |0i.
Going now to the entanglement properties presented
also in Section III.D, instantaneously to the above-
mentioned measures, the entangled qubits of Alice and
Bob, Aq2and Bq1will pass to the states |1iand |0iif
their pairs had finished in these states. (Remember the
properties of the Bell state β00 ).
A measure by Alice or Bob to these qubits will allow
to determine the GO or NO-GO condition of his part-
ner with probability 1, without the need of any classic
information exchange between them.
FIG. 7. Quantum Entanglement
E. Summary
Summarizing at the end of the process Alice and Bob
will have each of them the conditions GO or NO-
GOcorresponding to both partner positions.
So, two GO conditions will allow to settle from both
parties the Buy/Sell transaction making the correspond-
ing profit.
FIG. 8. Trade Summary
Taking in account that first measurement with the cho-
sen matrix is made at T1exact (∆T < 100 nanoseconds),
the second measurement would be made at T1+ 1 mi-
crosecond to have the assurance that the first measure-
ment has already taken place by the other party.
In this condition the trade is settle by both parties at
T1+2 microseconds using machines that perform Tick to
Trade in this 1 microsecond delay.
Trade is close and next cycle will start at T0+ 50 mil-
liseconds as the new T0.
To close this section let us, using the dynamics of the
market as presented in Section II, to make a quick and of
course not very serious calculation of the potential that
the above strategy could bring over this pair.
Considering that the GO conditions are established by
a minimum 1 pip (0.01%), price difference between the
positions, and these GO conditions have a probability
of 10% (minimum), the 20 Hz cycle gives us 2 trades
successful each second. Having 10 hours market trading,
he total number of successful trades are 72,000. With the
previous 1 pip profit per trade we obtain 7.2% profit per
day over the nominal amount of the transactions.
V. IMPLEMENTATION
A. Structure and Technologies
The previous devised strategy could be implemented
under the structure of a Hedge Fund (HF) under the cat-
egory of extremely technological High Frequency Trad-
ing.
The base as presented is the arbitration of one pair
(EUR / USD) at FOREX markets between New York
and London.
8
Technologies for the implementation will be forefront
in these three domains:
1. Tick-to-Trade
Tick-to-Trade in the order of 1 microsecond (1 µsec).
The current (posted) fastest time is 98 nanoseconds. So,
we can be 10 times slower to the top performers of the
table regarding this technology.
2. Communications
Communications between London and New York us-
ing existing Fibre-optics links or in a further step radio
waves over the Atlantic. Satellite connectivity with Low
Orbit Satellites could also be considered. In any case
our 43-msec. mark is easily achievable from any of these
technologies.
3. Quantum supremacy
Quantum supremacy in execution, through the entan-
glement mechanisms between the devices co-located with
the execution computers at the premises of the FOREX
markets in London and New York.
B. State of the Art
The markets of London and New York will be the
launching ones due to its maximum volume and relia-
bility. Also initially as described in the paper we can op-
erate over one pair. There’s no reason to think in further
expansions of the operations to other places and other
pairs, in which the HF can operate.
It’s clear that the hurdle for the expansion, and for the
launch of course, will be the availability of the devices
that will support the quantum supremacy.
As today, November 2019, the availability of qubits en-
tangled in different locations to perform quantum com-
munication is one of the hottest applications. The main
benefit for Quantum Communication is in the Quantum
Key Distribution (QKD) processes and currently is a
matter of normal business. In July last year, Alberto
Boaron of the University of Geneva, Switzerland, and
colleagues reported distributing secret keys using QKD
over a record distance of more than 400 kilometres of op-
tical fiber, at 6.5 kilobits per second. In contrast, com-
mercially available systems, such as the one sold by the
Geneva-based company ID Quantique, provide QKD over
50 kilometres of fibre.
The critical issue for all the current developments is
the distance that the photons as carriers of the quantum
information could achieve.
C. Disclaimer
The companies of the Author, RQuanTech (Geneva)
and Criptosasun (Madrid), are working in a device that
would implement the entanglement of qubits at the dis-
tances required by the setup of the methodology dis-
cussed in Section IV and the implementation presented
in Section V
Taking in account that this development is a work in
progress we can not now assure that the milestones pro-
posed will be achieved.
VI. CONCLUSIONS
It’s clear that Quantum Supremacy will not only dis-
rupts financial markets nor all the other markets in the
world.
Regarding the current quarrel between IBM and
Google over the claim of the later, Ethan Siegel has
written in Forbes magazine[9], “Progress in the world
of quantum computing is astounding, and despite the
claims of its detractors, systems with greater numbers of
qubits are undoubtedly on the horizon. When successful
quantum error-correction arrives (which will certainly re-
quire many more qubits and the necessity of addressing
and solving a number of other issues), we’ll be able to
extend the coherence timescale and perform even more
in-depth calculations”
ACKNOWLEDGMENTS
The Author wish to acknowledge the support of the
teams of RQuanTech and Criptosasun for their full sup-
port with the corrections and marvelous ideas.
[1] IBM, (2019), https://www.ibm.com/quantum-
computing/.
[2] I. L. Nielsen, Michael A.; Chuang, Quantum Computa-
tion and Quantum Information: 10th Anniversary Edition
(Cambridge University Press, 2010).
[3] J. A. Wheeler, Information, physics, quantum: the search
for links, in Complexity, entropy, and the physics of in-
formation (Westview Press, 1990).
[4] NASDAQ, (2019), https://www.nasdaq.com/glossary/h/high-
frequency-trading.
[5] D. Polen, Fix Globlal 2-12, 23 (2010).
[6] SEC, (2014), https://wallstreetonparade.com/2014/04/did-
the-sec-admit-that-it-knows-the-stock-market-is-rigged/.
[7] S. of SEC, Equity Market Structure Literature Review.
Part II: High Frequency Trading (U.S. Securities and Ex-
change Commission, 2014).
9
[8] R. S. M. Shorter, High Frequency Trading: Overview
of Recent Developments (Congressional Research Service,
2016).
[9] E. Siegel, Forbes Magazine 10-19, 27 (2019).