SlideShare una empresa de Scribd logo
1 de 8
arXiv:physics/0509256v130Sep2005 APS/123-QED
Analysis of Binarized High Frequency Financial Data
Naoya Sazuka∗
Corporate Human Resources Dept,
Sony Corporation,
4-10-18 Takanawa Minato-ku,
Tokyo, 108-0074 Japan
(Dated: October 3, 2005)
Abstract
A non-trivial probability structure is evident in the binary data extracted from the up/down price
movements of very high frequency data such as tick-by-tick data for USD/JPY. In this paper, we
analyze the Sony bank USD/JPY rates, ignoring the small deviations from the market price. We
then show there is a similar non-trivial probability structure in the Sony bank rate, in spite of the
Sony bank rate’s having less frequent and larger deviations than tick-by-tick data. However, this
probability structure is not found in the data which has been sampled from tick-by-tick data at
the same rate as the Sony bank rate. Therefore, the method of generating the Sony bank rate
from the market rate has the potential for practical use since the method retains the probability
structure as the sampling frequency decreases.
PACS numbers: 89.65.Gh
∗
Electronic address: Naoya.Sazuka@jp.sony.com
1
I. INTRODUCTION
In recent years, there have been various analyses of high frequency financial data which
has been within the past decade become available [1, 2]. High frequency financial data
describe the market behavior in fine detail and they have an interesting statistical property
that is not seen from the low frequency data.
In our previous works [3, 6], we showed the non-trivial probability structure in the tick-
by-tick data for USD/JPY with a focus on the direction of up/down price movements, as
opposed to the extent of price movement. A non-trivial probability structure is not apparent
from price changes themselves [4]. Such a probability structure is also observed in the high
frequency data of the NYSE for example GE data. In addition, we proposed the non-linear
logit model that agrees well with the empirical data.
The purpose of this paper is to understand in more detail the non-trivial probability
structure of binarized high frequency data. We analyze the Sony bank USD/JPY rates,
ignoring the small deviations from the market rate. We found that the non-trivial probability
structure is evident in the Sony bank USD/JPY rate, even if the data have about a 20 minute
interval between data points and larger price changes than tick-by-tick data.
The paper is organized as follows. In Section 2, we first explain the Sony bank and the
Sony bank rate. In Section 3, we analyze the Sony bank rate. We present our conclusions
in Section 4.
II. DATA
A. Sony bank
The Sony bank (http://moneykit.net/), launched in 2001, is an internet-based bank
for individual customers in Japan and aims to provide them with low cost services as close
as possible to the professional market environment. One of their main services is online
foreign-exchange trading, and this is quite popular. As of July 2005, the Sony bank deals
with eight currencies and they have about 130,000 foreign-currency accounts, which is about
30% of the total number of accounts. On the other hand, foreign-currency accounts are only
about 3% of all accounts in conventional banks.
There are three main reasons for the popularity of the bank’s foreign-exchange trading.
2
Firstly, trades can be made on the web 24 hours a day, 7 day a week. In conventional banks
in Japan, however, customers actually have to go to a branch and can trade only during
weekday business hours. Secondly, customers can trade at the Sony bank’s rate which at all
times closely reflects the market rate. In conventional banks, the trading rate for individual
customers is fixed at one time in the morning for the day’s trading. Finally, the bank’s low
transaction cost. For example, the trading cost of the Sony bank for USD/JPY is 0.25 yen
per dollar, which is about a quarter of the trading cost of conventional banks
B. Sony bank rate
Market rates
rate
tick
0.1 yen
0.1 yen
Sony bank rates
t0 t1
t2
Market rates
rate
tick
0.1 yen
0.1 yen
Sony bank rates
t0 t1
t2
FIG. 1: Update of the Sony bank rate for USD/JPY. The Sony bank USD/JPY rate is a solid line
and the market rate is a broken line. In this figure, the Sony bank USD/JPY rate is fixed from t0
to t1 and is updated to the market rate at t2, as the market rate exceeds the range of 0.1 yen at
t2 based on the market rate at t0.
In this paper, we analyze the Sony bank rate for USD/JPY. The Sony bank rate is the
foreign-currency exchange rate that customers having a foreign-currency account of the Sony
bank can deal at. The Sony bank rate is dependent on the market rate but is independent
of the customers’ orders. If the USD/JPY market rate changes more than 0.1 yen, the Sony
bank rate is updated to the market rate. Figure 1 is a schematic of the updating of the
Sony bank rate for USD/JPY. If the market rate changes along the broken line, the Sony
bank rate follows the solid line. Suppose the market rate (broken line) is the same as the
Sony bank rate (solid line) at the time t0. The Sony bank rate stays flat from t0 to t1, as
3
the market rate is in the range of 0.1 yen based on the market rate at t0. When the market
rate exceeds the range of 0.1 yen at t2, the Sony bank rate is updated to the market rate.
Since the Sony bank rate ignores small deviations in the market price by applying its
updating rule, the Sony bank rate has less frequent and larger price changes than tick-by-
tick data. The comparison is shown in Table I. The smallest price change, which is the
majority of the price changes, is 0.1 yen for the Sony bank rate. On the other hand, 40% of
price changes in tick-by-tick data are zero deviations between data points. We ignore such
cases in order to focus on only the direction of up/down price movements. Then tick-by-tick
data moves by a 0.01 yen unit.
TABLE I: The Sony bank USD/JPY rate and tick-by-tick data for USD/JPY
Sony bank rate tick-by-tick data
Number of data a day ∼70 ∼10,000
The smallest price change 0.1 yen 0.01 yen
Average interval between data ∼20 minutes ∼7 seconds
III. ANALYSIS
We analyze the Sony bank USD/JPY rates for the period of 22/10/2001 to 31/05/2005
(Figure 2). The data is composed of value S(t) of yen value per dollar at the Sony bank tick
step t. The number of data is 45,577.
As analyzed in our previous works [3, 6], we first binarize data extracting the direction
of the up/down price movement in the following way: X(t) = +1 if S(t + 1) − S(t) > 0 and
X(t) = −1 if S(t + 1) − S(t) < 0. According to the update rule, the Sony bank rate do not
have zero deviations between data points (S(t + 1) − S(t) = 0) not like tick-by-tick data.
Next we compute the conditional probabilities of binary data X(t) upto 4th order.
Then we show the non-trivial conditional probability structure of the binary data X(t).
Figure 3 is the probability structure of the Sony bank rates for whole period, and the first
half of the data and the second half of the data. The data was divided into two periods
to compare the tendencies. We show only “+” side of the conditional probabilities because
of the up/down symmetry. Although it would seem obvious that if the market is rising,
4
10000 20000 30000 40000
t ticks
110
115
120
125
130
135
S
¡t¢ Sony bank rates
FIG. 2: The Sony bank USD/JPY rates ploted against ticks for the period 22/10/2001 to
31/05/2005.
there must be more +’s than -’s, in fact, if we only have the data which is triggers and
change in the Sony bank rate, there will be over the long run a almost equal number
of + and - changes. This non-trivial probability structure is not only in the first order
conditional probabilities but also higher orders. Especially, it is likely to happen a regular
“zigzag” motion of alternating up and down, for example P(+|−, +, −, +) is 73%, rather
than irregular “zigzag” motions, for example P(+|−, +, −, +) is 63% and a motion in one
direction, for example P(+|+, +, +, +) is 40%.
It is interesting that the Sony bank rate has a probability structure, considering that
the average interval between data points is about 20 minutes. However, this probability
structure is not found in the data which has been sampled from tick-by-tick data at the
same rate as the Sony bank rate. Zhang looked similar higher order statistics and found
that the probability structure was not observed in the data recoded every 30 minutes [5].
In addition, the probability structure of the Sony bank rate is not the effect of the negative
first-order auto correlation in the second scale tick-by-tick data, since many transactions
occur during 20 minutes. Therefore, the method of generating the Sony bank rate from the
market rate can retain the probability structure as a sampling frequency decreases.
The probability structure of the Sony bank rate is not only based on less frequent sampling
but also larger deviations than tick-by-tick data. As shown in an Table I, the probability
structure of tick-by-tick data is mostly 0.01 yen price changes within 7 seconds on an average
5
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
P(+)
P(+|+)
P(+|+,+)
P(+|-,+)
P(+|+,+,+)
P(+|-,+,+)
P(+|+,-,+)
P(+|+,+,-)
P(+|+,+,+,+)
P(+|-,+,+,+)
P(+|+,-,+,+)
P(+|+,+,-,+)
P(+|+,+,+,-)
P(+|-,-,+,+)
P(+|-,+,-,+)
P(+|-,+,+,-)
all data
1st half
2nd half
FIG. 3: Probability structure of the Sony bank USD/JPY rate. Each conditional probability has
three bars. The upper one is the result of whole period 22/10/2001 to 31/05/2005. The middle is
for the first half of the data. The lower is for the second half of the data. The error bars represent
a 95% confidence interval on the mean
and sometimes even shorter. This may be too frequent and small for practical use. However,
the Sony bank rate essentially reflects mostly 0.1 yen price changes and about 20 minutes
interval between price chages. Therefore, there is a better chance to take advantage of the
Sony bank rate.
In addition, we compare the probability structure of GE stock data on NYSE shown
in [6]. The data was collected in December 2002. The number of GE data for one
day is about 180,000, which is more frequent than the tick-by-tick data for USD/JPY.
Figure 4 shows that the reliability of the probability structure increases as the sam-
pling frequency increases, comparing tick-by-tick data for USD/JPY, the Sony bank
USD/JPY rate and GE data on NYSE. The exceptions are regular “zigzag” motions such as
P(+|−, +), P(+|+, −, +), P(+|−, +, −, +). In these cases, the probability structure of the
Sony bank rate is more reliable than the tick-by-tick USD/JPY rate.
6
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
P(+)
P(+|+)
P(+|+,+)
P(+|-,+)
P(+|+,+,+)
P(+|-,+,+)
P(+|+,-,+)
P(+|+,+,-)
P(+|+,+,+,+)
P(+|-,+,+,+)
P(+|+,-,+,+)
P(+|+,+,-,+)
P(+|+,+,+,-)
P(+|-,-,+,+)
P(+|-,+,-,+)
P(+|-,+,+,-)
Sony bank rate
tick-by-tick data
GE
FIG. 4: Comparison of the bias among the tick-by-tick data for GE on NYSE, tick-by-tick data for
USD/JPY and Sony bank USD/JPY rates. The reliability of the probability structure increaces
as the frequency of the data increaces. The error bars represent a 95% confidence interval on the
mean.
IV. CONCLUSION
We analyzed the Sony bank USD/JPY rate, which closely reflects the market rate, with
a focus on the up/down price movement, as opposed to the extent of price movement. The
Sony bank has a lower sampling rate and shows larger price change deviations than the
market rate. The main result is that a probability structure is observed in the Sony bank
rate, in spite of the interval between data points of the Sony bank rate is about 20 minutes.
However, this probability structure is not found in the data which has been sampled from
tick-by-tick data at the same rate as the Sony bank rate. Therefore, the method of generating
the Sony bank rate from the market rate can retain the probability structure as the sampling
frequency decreases. There is a better chance to take advantage of the Sony bank rate in
practical use.
7
Acknowledgments
I would like to thank Shigeru Ishi, President of the Sony bank, for kindly providing the
Sony bank data. The NYSE data are from King’s College London.
[1] M. M. Dacorogna, R.Gencay, U. A. Muller, R. B. Olsen, V. Pictet, in An Introduction to
High-Frequency Finance (Academic Press, San Diego, 2001).
[2] H. Takayasu (Ed), Empirical Science of Financial Fluctuations:The Advent of Econophysics. ,
(Springer-Verlag Tokyo, 2002).
[3] T. Ohira, N. Sazuka, K. Marumo, T. Shimizu, M. Takayasu and H. Takayasu, Physica A, 308,
N1-4, pp. 368–374, (2002).
[4] N. Sazuka, T. Ohira, K. Marumo, T. Shimizu, M. Takayasu and H. Takayasu, Physica A, 324,
N1-2, pp. 366–371, (2003).
[5] Y. Zhang, Physica A 269, pp. 30–44, (1999).
[6] N. Sazuka, Physica A, 355, pp. 183–189, (2005).
8

Más contenido relacionado

Similar a An binarized fin data

Collateralised swap agreements
Collateralised swap agreementsCollateralised swap agreements
Collateralised swap agreementstheo1305
 
ANALYSIS OF INTRADAY TRADING OF INDEX OPTION IN KOREAN OPTION MARKET
ANALYSIS OF INTRADAY TRADING OF INDEX OPTION IN KOREAN OPTION MARKETANALYSIS OF INTRADAY TRADING OF INDEX OPTION IN KOREAN OPTION MARKET
ANALYSIS OF INTRADAY TRADING OF INDEX OPTION IN KOREAN OPTION MARKETcscpconf
 
How to Apply VECM to Detect the J-Curve in the Case of Japan
How to Apply VECM to Detect the J-Curve in the Case of JapanHow to Apply VECM to Detect the J-Curve in the Case of Japan
How to Apply VECM to Detect the J-Curve in the Case of JapanDr. Kelly YiYu Lin
 
Analysis of intraday trading of index option in korean option market
Analysis of intraday trading of index option in korean option marketAnalysis of intraday trading of index option in korean option market
Analysis of intraday trading of index option in korean option marketcsandit
 
Determination trading puchong
Determination trading puchongDetermination trading puchong
Determination trading puchongReza Sadeghdoost
 
Market focus inflation_2017
Market  focus inflation_2017Market  focus inflation_2017
Market focus inflation_2017Cristiana Corno
 
CENTRAL BANK INTERVENTION AND EXCHANGE RATE VOLATILITY, ITS CONTINUOUS AND JU...
CENTRAL BANK INTERVENTION AND EXCHANGE RATE VOLATILITY, ITS CONTINUOUS AND JU...CENTRAL BANK INTERVENTION AND EXCHANGE RATE VOLATILITY, ITS CONTINUOUS AND JU...
CENTRAL BANK INTERVENTION AND EXCHANGE RATE VOLATILITY, ITS CONTINUOUS AND JU...Nicha Tatsaneeyapan
 
Finance Time Series slides.pptx
Finance Time Series slides.pptxFinance Time Series slides.pptx
Finance Time Series slides.pptxdsaf6
 
Keynesian Model With Interest
Keynesian Model With InterestKeynesian Model With Interest
Keynesian Model With InterestSaurabh Goel
 
Japan's Abenomics and its implication on Asian economy
Japan's Abenomics and its implication on Asian economy Japan's Abenomics and its implication on Asian economy
Japan's Abenomics and its implication on Asian economy London Business School
 
Science in the City 2022 Prof J Jamesv[2022]
Science in the City 2022 Prof J Jamesv[2022]Science in the City 2022 Prof J Jamesv[2022]
Science in the City 2022 Prof J Jamesv[2022]ikhmalsyuqkinbinisma
 
W K11 International Monetary System
W K11 International  Monetary SystemW K11 International  Monetary System
W K11 International Monetary SystemAjOb
 
ICICI Prudential Floating Interest Fund
ICICI Prudential Floating Interest FundICICI Prudential Floating Interest Fund
ICICI Prudential Floating Interest Fundiciciprumf
 
POST CRISIS STUDY OF THE UK BANKING SYSTEM using the Rose and Panzar H statistic
POST CRISIS STUDY OF THE UK BANKING SYSTEM using the Rose and Panzar H statisticPOST CRISIS STUDY OF THE UK BANKING SYSTEM using the Rose and Panzar H statistic
POST CRISIS STUDY OF THE UK BANKING SYSTEM using the Rose and Panzar H statisticSubhrodip Sengupta
 
The Mortgage Dilemma - which rate type?
The Mortgage Dilemma - which rate type?The Mortgage Dilemma - which rate type?
The Mortgage Dilemma - which rate type?riskxego
 
Study of volatility in foreign exchange market
Study of volatility in foreign exchange marketStudy of volatility in foreign exchange market
Study of volatility in foreign exchange marketSunita Jindal
 

Similar a An binarized fin data (20)

Collateralised swap agreements
Collateralised swap agreementsCollateralised swap agreements
Collateralised swap agreements
 
ANALYSIS OF INTRADAY TRADING OF INDEX OPTION IN KOREAN OPTION MARKET
ANALYSIS OF INTRADAY TRADING OF INDEX OPTION IN KOREAN OPTION MARKETANALYSIS OF INTRADAY TRADING OF INDEX OPTION IN KOREAN OPTION MARKET
ANALYSIS OF INTRADAY TRADING OF INDEX OPTION IN KOREAN OPTION MARKET
 
How to Apply VECM to Detect the J-Curve in the Case of Japan
How to Apply VECM to Detect the J-Curve in the Case of JapanHow to Apply VECM to Detect the J-Curve in the Case of Japan
How to Apply VECM to Detect the J-Curve in the Case of Japan
 
Analysis of intraday trading of index option in korean option market
Analysis of intraday trading of index option in korean option marketAnalysis of intraday trading of index option in korean option market
Analysis of intraday trading of index option in korean option market
 
Determination trading puchong
Determination trading puchongDetermination trading puchong
Determination trading puchong
 
Journal
JournalJournal
Journal
 
Market focus inflation_2017
Market  focus inflation_2017Market  focus inflation_2017
Market focus inflation_2017
 
CENTRAL BANK INTERVENTION AND EXCHANGE RATE VOLATILITY, ITS CONTINUOUS AND JU...
CENTRAL BANK INTERVENTION AND EXCHANGE RATE VOLATILITY, ITS CONTINUOUS AND JU...CENTRAL BANK INTERVENTION AND EXCHANGE RATE VOLATILITY, ITS CONTINUOUS AND JU...
CENTRAL BANK INTERVENTION AND EXCHANGE RATE VOLATILITY, ITS CONTINUOUS AND JU...
 
Tianjin Case Study
Tianjin Case StudyTianjin Case Study
Tianjin Case Study
 
Finance Time Series slides.pptx
Finance Time Series slides.pptxFinance Time Series slides.pptx
Finance Time Series slides.pptx
 
Keynesian Model With Interest
Keynesian Model With InterestKeynesian Model With Interest
Keynesian Model With Interest
 
Japan's Abenomics and its implication on Asian economy
Japan's Abenomics and its implication on Asian economy Japan's Abenomics and its implication on Asian economy
Japan's Abenomics and its implication on Asian economy
 
Science in the City 2022 Prof J Jamesv[2022]
Science in the City 2022 Prof J Jamesv[2022]Science in the City 2022 Prof J Jamesv[2022]
Science in the City 2022 Prof J Jamesv[2022]
 
W K11 International Monetary System
W K11 International  Monetary SystemW K11 International  Monetary System
W K11 International Monetary System
 
ICICI Prudential Floating Interest Fund
ICICI Prudential Floating Interest FundICICI Prudential Floating Interest Fund
ICICI Prudential Floating Interest Fund
 
Real er determinants
Real er determinantsReal er determinants
Real er determinants
 
Volatility, Firm Size and Economic Growth: Evidence from Chinese Stock Market
Volatility, Firm Size and Economic Growth: Evidence from Chinese Stock MarketVolatility, Firm Size and Economic Growth: Evidence from Chinese Stock Market
Volatility, Firm Size and Economic Growth: Evidence from Chinese Stock Market
 
POST CRISIS STUDY OF THE UK BANKING SYSTEM using the Rose and Panzar H statistic
POST CRISIS STUDY OF THE UK BANKING SYSTEM using the Rose and Panzar H statisticPOST CRISIS STUDY OF THE UK BANKING SYSTEM using the Rose and Panzar H statistic
POST CRISIS STUDY OF THE UK BANKING SYSTEM using the Rose and Panzar H statistic
 
The Mortgage Dilemma - which rate type?
The Mortgage Dilemma - which rate type?The Mortgage Dilemma - which rate type?
The Mortgage Dilemma - which rate type?
 
Study of volatility in foreign exchange market
Study of volatility in foreign exchange marketStudy of volatility in foreign exchange market
Study of volatility in foreign exchange market
 

Más de ivan weinel

Applyng wtd pr cttn nets
Applyng wtd pr cttn netsApplyng wtd pr cttn nets
Applyng wtd pr cttn netsivan weinel
 
Strumsky lobo (2011) does patenting intensity beget quality
Strumsky lobo (2011) does patenting intensity beget qualityStrumsky lobo (2011) does patenting intensity beget quality
Strumsky lobo (2011) does patenting intensity beget qualityivan weinel
 
Fr1063 park-moving-average-ratio-and-momentum
Fr1063 park-moving-average-ratio-and-momentumFr1063 park-moving-average-ratio-and-momentum
Fr1063 park-moving-average-ratio-and-momentumivan weinel
 
Textual analysis of stock market
Textual analysis of stock marketTextual analysis of stock market
Textual analysis of stock marketivan weinel
 
Simple regs rdc ineq
Simple regs rdc ineqSimple regs rdc ineq
Simple regs rdc ineqivan weinel
 
Mass exrgy effcncy us ecnmy ayres
Mass exrgy effcncy us ecnmy ayresMass exrgy effcncy us ecnmy ayres
Mass exrgy effcncy us ecnmy ayresivan weinel
 
Artfcl tchncl an
Artfcl tchncl anArtfcl tchncl an
Artfcl tchncl anivan weinel
 
Ultimatevaluefinder sampleissue
Ultimatevaluefinder sampleissueUltimatevaluefinder sampleissue
Ultimatevaluefinder sampleissueivan weinel
 
System dynamics review
System dynamics reviewSystem dynamics review
System dynamics reviewivan weinel
 

Más de ivan weinel (11)

Applyng wtd pr cttn nets
Applyng wtd pr cttn netsApplyng wtd pr cttn nets
Applyng wtd pr cttn nets
 
Strumsky lobo (2011) does patenting intensity beget quality
Strumsky lobo (2011) does patenting intensity beget qualityStrumsky lobo (2011) does patenting intensity beget quality
Strumsky lobo (2011) does patenting intensity beget quality
 
Fr1063 park-moving-average-ratio-and-momentum
Fr1063 park-moving-average-ratio-and-momentumFr1063 park-moving-average-ratio-and-momentum
Fr1063 park-moving-average-ratio-and-momentum
 
Textual analysis of stock market
Textual analysis of stock marketTextual analysis of stock market
Textual analysis of stock market
 
Simple regs rdc ineq
Simple regs rdc ineqSimple regs rdc ineq
Simple regs rdc ineq
 
Sk rndm grmmrs
Sk rndm grmmrsSk rndm grmmrs
Sk rndm grmmrs
 
Mass exrgy effcncy us ecnmy ayres
Mass exrgy effcncy us ecnmy ayresMass exrgy effcncy us ecnmy ayres
Mass exrgy effcncy us ecnmy ayres
 
Mngng rsk mmntm
Mngng rsk mmntmMngng rsk mmntm
Mngng rsk mmntm
 
Artfcl tchncl an
Artfcl tchncl anArtfcl tchncl an
Artfcl tchncl an
 
Ultimatevaluefinder sampleissue
Ultimatevaluefinder sampleissueUltimatevaluefinder sampleissue
Ultimatevaluefinder sampleissue
 
System dynamics review
System dynamics reviewSystem dynamics review
System dynamics review
 

Último

Best VIP Call Girls Noida Sector 18 Call Me: 8448380779
Best VIP Call Girls Noida Sector 18 Call Me: 8448380779Best VIP Call Girls Noida Sector 18 Call Me: 8448380779
Best VIP Call Girls Noida Sector 18 Call Me: 8448380779Delhi Call girls
 
Solution Manual for Financial Accounting, 11th Edition by Robert Libby, Patri...
Solution Manual for Financial Accounting, 11th Edition by Robert Libby, Patri...Solution Manual for Financial Accounting, 11th Edition by Robert Libby, Patri...
Solution Manual for Financial Accounting, 11th Edition by Robert Libby, Patri...ssifa0344
 
Pooja 9892124323 : Call Girl in Juhu Escorts Service Free Home Delivery
Pooja 9892124323 : Call Girl in Juhu Escorts Service Free Home DeliveryPooja 9892124323 : Call Girl in Juhu Escorts Service Free Home Delivery
Pooja 9892124323 : Call Girl in Juhu Escorts Service Free Home DeliveryPooja Nehwal
 
Log your LOA pain with Pension Lab's brilliant campaign
Log your LOA pain with Pension Lab's brilliant campaignLog your LOA pain with Pension Lab's brilliant campaign
Log your LOA pain with Pension Lab's brilliant campaignHenry Tapper
 
06_Joeri Van Speybroek_Dell_MeetupDora&Cybersecurity.pdf
06_Joeri Van Speybroek_Dell_MeetupDora&Cybersecurity.pdf06_Joeri Van Speybroek_Dell_MeetupDora&Cybersecurity.pdf
06_Joeri Van Speybroek_Dell_MeetupDora&Cybersecurity.pdfFinTech Belgium
 
High Class Call Girls Nashik Maya 7001305949 Independent Escort Service Nashik
High Class Call Girls Nashik Maya 7001305949 Independent Escort Service NashikHigh Class Call Girls Nashik Maya 7001305949 Independent Escort Service Nashik
High Class Call Girls Nashik Maya 7001305949 Independent Escort Service NashikCall Girls in Nagpur High Profile
 
00_Main ppt_MeetupDORA&CyberSecurity.pptx
00_Main ppt_MeetupDORA&CyberSecurity.pptx00_Main ppt_MeetupDORA&CyberSecurity.pptx
00_Main ppt_MeetupDORA&CyberSecurity.pptxFinTech Belgium
 
The Economic History of the U.S. Lecture 17.pdf
The Economic History of the U.S. Lecture 17.pdfThe Economic History of the U.S. Lecture 17.pdf
The Economic History of the U.S. Lecture 17.pdfGale Pooley
 
The Economic History of the U.S. Lecture 25.pdf
The Economic History of the U.S. Lecture 25.pdfThe Economic History of the U.S. Lecture 25.pdf
The Economic History of the U.S. Lecture 25.pdfGale Pooley
 
03_Emmanuel Ndiaye_Degroof Petercam.pptx
03_Emmanuel Ndiaye_Degroof Petercam.pptx03_Emmanuel Ndiaye_Degroof Petercam.pptx
03_Emmanuel Ndiaye_Degroof Petercam.pptxFinTech Belgium
 
Vip Call US 📞 7738631006 ✅Call Girls In Sakinaka ( Mumbai )
Vip Call US 📞 7738631006 ✅Call Girls In Sakinaka ( Mumbai )Vip Call US 📞 7738631006 ✅Call Girls In Sakinaka ( Mumbai )
Vip Call US 📞 7738631006 ✅Call Girls In Sakinaka ( Mumbai )Pooja Nehwal
 
VIP Call Girl in Mira Road 💧 9920725232 ( Call Me ) Get A New Crush Everyday ...
VIP Call Girl in Mira Road 💧 9920725232 ( Call Me ) Get A New Crush Everyday ...VIP Call Girl in Mira Road 💧 9920725232 ( Call Me ) Get A New Crush Everyday ...
VIP Call Girl in Mira Road 💧 9920725232 ( Call Me ) Get A New Crush Everyday ...dipikadinghjn ( Why You Choose Us? ) Escorts
 
Stock Market Brief Deck (Under Pressure).pdf
Stock Market Brief Deck (Under Pressure).pdfStock Market Brief Deck (Under Pressure).pdf
Stock Market Brief Deck (Under Pressure).pdfMichael Silva
 
High Class Call Girls Nagpur Grishma Call 7001035870 Meet With Nagpur Escorts
High Class Call Girls Nagpur Grishma Call 7001035870 Meet With Nagpur EscortsHigh Class Call Girls Nagpur Grishma Call 7001035870 Meet With Nagpur Escorts
High Class Call Girls Nagpur Grishma Call 7001035870 Meet With Nagpur Escortsranjana rawat
 
The Economic History of the U.S. Lecture 22.pdf
The Economic History of the U.S. Lecture 22.pdfThe Economic History of the U.S. Lecture 22.pdf
The Economic History of the U.S. Lecture 22.pdfGale Pooley
 
Booking open Available Pune Call Girls Wadgaon Sheri 6297143586 Call Hot Ind...
Booking open Available Pune Call Girls Wadgaon Sheri  6297143586 Call Hot Ind...Booking open Available Pune Call Girls Wadgaon Sheri  6297143586 Call Hot Ind...
Booking open Available Pune Call Girls Wadgaon Sheri 6297143586 Call Hot Ind...Call Girls in Nagpur High Profile
 
The Economic History of the U.S. Lecture 21.pdf
The Economic History of the U.S. Lecture 21.pdfThe Economic History of the U.S. Lecture 21.pdf
The Economic History of the U.S. Lecture 21.pdfGale Pooley
 
The Economic History of the U.S. Lecture 20.pdf
The Economic History of the U.S. Lecture 20.pdfThe Economic History of the U.S. Lecture 20.pdf
The Economic History of the U.S. Lecture 20.pdfGale Pooley
 
Independent Call Girl Number in Kurla Mumbai📲 Pooja Nehwal 9892124323 💞 Full ...
Independent Call Girl Number in Kurla Mumbai📲 Pooja Nehwal 9892124323 💞 Full ...Independent Call Girl Number in Kurla Mumbai📲 Pooja Nehwal 9892124323 💞 Full ...
Independent Call Girl Number in Kurla Mumbai📲 Pooja Nehwal 9892124323 💞 Full ...Pooja Nehwal
 
Top Rated Pune Call Girls Viman Nagar ⟟ 6297143586 ⟟ Call Me For Genuine Sex...
Top Rated  Pune Call Girls Viman Nagar ⟟ 6297143586 ⟟ Call Me For Genuine Sex...Top Rated  Pune Call Girls Viman Nagar ⟟ 6297143586 ⟟ Call Me For Genuine Sex...
Top Rated Pune Call Girls Viman Nagar ⟟ 6297143586 ⟟ Call Me For Genuine Sex...Call Girls in Nagpur High Profile
 

Último (20)

Best VIP Call Girls Noida Sector 18 Call Me: 8448380779
Best VIP Call Girls Noida Sector 18 Call Me: 8448380779Best VIP Call Girls Noida Sector 18 Call Me: 8448380779
Best VIP Call Girls Noida Sector 18 Call Me: 8448380779
 
Solution Manual for Financial Accounting, 11th Edition by Robert Libby, Patri...
Solution Manual for Financial Accounting, 11th Edition by Robert Libby, Patri...Solution Manual for Financial Accounting, 11th Edition by Robert Libby, Patri...
Solution Manual for Financial Accounting, 11th Edition by Robert Libby, Patri...
 
Pooja 9892124323 : Call Girl in Juhu Escorts Service Free Home Delivery
Pooja 9892124323 : Call Girl in Juhu Escorts Service Free Home DeliveryPooja 9892124323 : Call Girl in Juhu Escorts Service Free Home Delivery
Pooja 9892124323 : Call Girl in Juhu Escorts Service Free Home Delivery
 
Log your LOA pain with Pension Lab's brilliant campaign
Log your LOA pain with Pension Lab's brilliant campaignLog your LOA pain with Pension Lab's brilliant campaign
Log your LOA pain with Pension Lab's brilliant campaign
 
06_Joeri Van Speybroek_Dell_MeetupDora&Cybersecurity.pdf
06_Joeri Van Speybroek_Dell_MeetupDora&Cybersecurity.pdf06_Joeri Van Speybroek_Dell_MeetupDora&Cybersecurity.pdf
06_Joeri Van Speybroek_Dell_MeetupDora&Cybersecurity.pdf
 
High Class Call Girls Nashik Maya 7001305949 Independent Escort Service Nashik
High Class Call Girls Nashik Maya 7001305949 Independent Escort Service NashikHigh Class Call Girls Nashik Maya 7001305949 Independent Escort Service Nashik
High Class Call Girls Nashik Maya 7001305949 Independent Escort Service Nashik
 
00_Main ppt_MeetupDORA&CyberSecurity.pptx
00_Main ppt_MeetupDORA&CyberSecurity.pptx00_Main ppt_MeetupDORA&CyberSecurity.pptx
00_Main ppt_MeetupDORA&CyberSecurity.pptx
 
The Economic History of the U.S. Lecture 17.pdf
The Economic History of the U.S. Lecture 17.pdfThe Economic History of the U.S. Lecture 17.pdf
The Economic History of the U.S. Lecture 17.pdf
 
The Economic History of the U.S. Lecture 25.pdf
The Economic History of the U.S. Lecture 25.pdfThe Economic History of the U.S. Lecture 25.pdf
The Economic History of the U.S. Lecture 25.pdf
 
03_Emmanuel Ndiaye_Degroof Petercam.pptx
03_Emmanuel Ndiaye_Degroof Petercam.pptx03_Emmanuel Ndiaye_Degroof Petercam.pptx
03_Emmanuel Ndiaye_Degroof Petercam.pptx
 
Vip Call US 📞 7738631006 ✅Call Girls In Sakinaka ( Mumbai )
Vip Call US 📞 7738631006 ✅Call Girls In Sakinaka ( Mumbai )Vip Call US 📞 7738631006 ✅Call Girls In Sakinaka ( Mumbai )
Vip Call US 📞 7738631006 ✅Call Girls In Sakinaka ( Mumbai )
 
VIP Call Girl in Mira Road 💧 9920725232 ( Call Me ) Get A New Crush Everyday ...
VIP Call Girl in Mira Road 💧 9920725232 ( Call Me ) Get A New Crush Everyday ...VIP Call Girl in Mira Road 💧 9920725232 ( Call Me ) Get A New Crush Everyday ...
VIP Call Girl in Mira Road 💧 9920725232 ( Call Me ) Get A New Crush Everyday ...
 
Stock Market Brief Deck (Under Pressure).pdf
Stock Market Brief Deck (Under Pressure).pdfStock Market Brief Deck (Under Pressure).pdf
Stock Market Brief Deck (Under Pressure).pdf
 
High Class Call Girls Nagpur Grishma Call 7001035870 Meet With Nagpur Escorts
High Class Call Girls Nagpur Grishma Call 7001035870 Meet With Nagpur EscortsHigh Class Call Girls Nagpur Grishma Call 7001035870 Meet With Nagpur Escorts
High Class Call Girls Nagpur Grishma Call 7001035870 Meet With Nagpur Escorts
 
The Economic History of the U.S. Lecture 22.pdf
The Economic History of the U.S. Lecture 22.pdfThe Economic History of the U.S. Lecture 22.pdf
The Economic History of the U.S. Lecture 22.pdf
 
Booking open Available Pune Call Girls Wadgaon Sheri 6297143586 Call Hot Ind...
Booking open Available Pune Call Girls Wadgaon Sheri  6297143586 Call Hot Ind...Booking open Available Pune Call Girls Wadgaon Sheri  6297143586 Call Hot Ind...
Booking open Available Pune Call Girls Wadgaon Sheri 6297143586 Call Hot Ind...
 
The Economic History of the U.S. Lecture 21.pdf
The Economic History of the U.S. Lecture 21.pdfThe Economic History of the U.S. Lecture 21.pdf
The Economic History of the U.S. Lecture 21.pdf
 
The Economic History of the U.S. Lecture 20.pdf
The Economic History of the U.S. Lecture 20.pdfThe Economic History of the U.S. Lecture 20.pdf
The Economic History of the U.S. Lecture 20.pdf
 
Independent Call Girl Number in Kurla Mumbai📲 Pooja Nehwal 9892124323 💞 Full ...
Independent Call Girl Number in Kurla Mumbai📲 Pooja Nehwal 9892124323 💞 Full ...Independent Call Girl Number in Kurla Mumbai📲 Pooja Nehwal 9892124323 💞 Full ...
Independent Call Girl Number in Kurla Mumbai📲 Pooja Nehwal 9892124323 💞 Full ...
 
Top Rated Pune Call Girls Viman Nagar ⟟ 6297143586 ⟟ Call Me For Genuine Sex...
Top Rated  Pune Call Girls Viman Nagar ⟟ 6297143586 ⟟ Call Me For Genuine Sex...Top Rated  Pune Call Girls Viman Nagar ⟟ 6297143586 ⟟ Call Me For Genuine Sex...
Top Rated Pune Call Girls Viman Nagar ⟟ 6297143586 ⟟ Call Me For Genuine Sex...
 

An binarized fin data

  • 1. arXiv:physics/0509256v130Sep2005 APS/123-QED Analysis of Binarized High Frequency Financial Data Naoya Sazuka∗ Corporate Human Resources Dept, Sony Corporation, 4-10-18 Takanawa Minato-ku, Tokyo, 108-0074 Japan (Dated: October 3, 2005) Abstract A non-trivial probability structure is evident in the binary data extracted from the up/down price movements of very high frequency data such as tick-by-tick data for USD/JPY. In this paper, we analyze the Sony bank USD/JPY rates, ignoring the small deviations from the market price. We then show there is a similar non-trivial probability structure in the Sony bank rate, in spite of the Sony bank rate’s having less frequent and larger deviations than tick-by-tick data. However, this probability structure is not found in the data which has been sampled from tick-by-tick data at the same rate as the Sony bank rate. Therefore, the method of generating the Sony bank rate from the market rate has the potential for practical use since the method retains the probability structure as the sampling frequency decreases. PACS numbers: 89.65.Gh ∗ Electronic address: Naoya.Sazuka@jp.sony.com 1
  • 2. I. INTRODUCTION In recent years, there have been various analyses of high frequency financial data which has been within the past decade become available [1, 2]. High frequency financial data describe the market behavior in fine detail and they have an interesting statistical property that is not seen from the low frequency data. In our previous works [3, 6], we showed the non-trivial probability structure in the tick- by-tick data for USD/JPY with a focus on the direction of up/down price movements, as opposed to the extent of price movement. A non-trivial probability structure is not apparent from price changes themselves [4]. Such a probability structure is also observed in the high frequency data of the NYSE for example GE data. In addition, we proposed the non-linear logit model that agrees well with the empirical data. The purpose of this paper is to understand in more detail the non-trivial probability structure of binarized high frequency data. We analyze the Sony bank USD/JPY rates, ignoring the small deviations from the market rate. We found that the non-trivial probability structure is evident in the Sony bank USD/JPY rate, even if the data have about a 20 minute interval between data points and larger price changes than tick-by-tick data. The paper is organized as follows. In Section 2, we first explain the Sony bank and the Sony bank rate. In Section 3, we analyze the Sony bank rate. We present our conclusions in Section 4. II. DATA A. Sony bank The Sony bank (http://moneykit.net/), launched in 2001, is an internet-based bank for individual customers in Japan and aims to provide them with low cost services as close as possible to the professional market environment. One of their main services is online foreign-exchange trading, and this is quite popular. As of July 2005, the Sony bank deals with eight currencies and they have about 130,000 foreign-currency accounts, which is about 30% of the total number of accounts. On the other hand, foreign-currency accounts are only about 3% of all accounts in conventional banks. There are three main reasons for the popularity of the bank’s foreign-exchange trading. 2
  • 3. Firstly, trades can be made on the web 24 hours a day, 7 day a week. In conventional banks in Japan, however, customers actually have to go to a branch and can trade only during weekday business hours. Secondly, customers can trade at the Sony bank’s rate which at all times closely reflects the market rate. In conventional banks, the trading rate for individual customers is fixed at one time in the morning for the day’s trading. Finally, the bank’s low transaction cost. For example, the trading cost of the Sony bank for USD/JPY is 0.25 yen per dollar, which is about a quarter of the trading cost of conventional banks B. Sony bank rate Market rates rate tick 0.1 yen 0.1 yen Sony bank rates t0 t1 t2 Market rates rate tick 0.1 yen 0.1 yen Sony bank rates t0 t1 t2 FIG. 1: Update of the Sony bank rate for USD/JPY. The Sony bank USD/JPY rate is a solid line and the market rate is a broken line. In this figure, the Sony bank USD/JPY rate is fixed from t0 to t1 and is updated to the market rate at t2, as the market rate exceeds the range of 0.1 yen at t2 based on the market rate at t0. In this paper, we analyze the Sony bank rate for USD/JPY. The Sony bank rate is the foreign-currency exchange rate that customers having a foreign-currency account of the Sony bank can deal at. The Sony bank rate is dependent on the market rate but is independent of the customers’ orders. If the USD/JPY market rate changes more than 0.1 yen, the Sony bank rate is updated to the market rate. Figure 1 is a schematic of the updating of the Sony bank rate for USD/JPY. If the market rate changes along the broken line, the Sony bank rate follows the solid line. Suppose the market rate (broken line) is the same as the Sony bank rate (solid line) at the time t0. The Sony bank rate stays flat from t0 to t1, as 3
  • 4. the market rate is in the range of 0.1 yen based on the market rate at t0. When the market rate exceeds the range of 0.1 yen at t2, the Sony bank rate is updated to the market rate. Since the Sony bank rate ignores small deviations in the market price by applying its updating rule, the Sony bank rate has less frequent and larger price changes than tick-by- tick data. The comparison is shown in Table I. The smallest price change, which is the majority of the price changes, is 0.1 yen for the Sony bank rate. On the other hand, 40% of price changes in tick-by-tick data are zero deviations between data points. We ignore such cases in order to focus on only the direction of up/down price movements. Then tick-by-tick data moves by a 0.01 yen unit. TABLE I: The Sony bank USD/JPY rate and tick-by-tick data for USD/JPY Sony bank rate tick-by-tick data Number of data a day ∼70 ∼10,000 The smallest price change 0.1 yen 0.01 yen Average interval between data ∼20 minutes ∼7 seconds III. ANALYSIS We analyze the Sony bank USD/JPY rates for the period of 22/10/2001 to 31/05/2005 (Figure 2). The data is composed of value S(t) of yen value per dollar at the Sony bank tick step t. The number of data is 45,577. As analyzed in our previous works [3, 6], we first binarize data extracting the direction of the up/down price movement in the following way: X(t) = +1 if S(t + 1) − S(t) > 0 and X(t) = −1 if S(t + 1) − S(t) < 0. According to the update rule, the Sony bank rate do not have zero deviations between data points (S(t + 1) − S(t) = 0) not like tick-by-tick data. Next we compute the conditional probabilities of binary data X(t) upto 4th order. Then we show the non-trivial conditional probability structure of the binary data X(t). Figure 3 is the probability structure of the Sony bank rates for whole period, and the first half of the data and the second half of the data. The data was divided into two periods to compare the tendencies. We show only “+” side of the conditional probabilities because of the up/down symmetry. Although it would seem obvious that if the market is rising, 4
  • 5. 10000 20000 30000 40000 t ticks 110 115 120 125 130 135 S ¡t¢ Sony bank rates FIG. 2: The Sony bank USD/JPY rates ploted against ticks for the period 22/10/2001 to 31/05/2005. there must be more +’s than -’s, in fact, if we only have the data which is triggers and change in the Sony bank rate, there will be over the long run a almost equal number of + and - changes. This non-trivial probability structure is not only in the first order conditional probabilities but also higher orders. Especially, it is likely to happen a regular “zigzag” motion of alternating up and down, for example P(+|−, +, −, +) is 73%, rather than irregular “zigzag” motions, for example P(+|−, +, −, +) is 63% and a motion in one direction, for example P(+|+, +, +, +) is 40%. It is interesting that the Sony bank rate has a probability structure, considering that the average interval between data points is about 20 minutes. However, this probability structure is not found in the data which has been sampled from tick-by-tick data at the same rate as the Sony bank rate. Zhang looked similar higher order statistics and found that the probability structure was not observed in the data recoded every 30 minutes [5]. In addition, the probability structure of the Sony bank rate is not the effect of the negative first-order auto correlation in the second scale tick-by-tick data, since many transactions occur during 20 minutes. Therefore, the method of generating the Sony bank rate from the market rate can retain the probability structure as a sampling frequency decreases. The probability structure of the Sony bank rate is not only based on less frequent sampling but also larger deviations than tick-by-tick data. As shown in an Table I, the probability structure of tick-by-tick data is mostly 0.01 yen price changes within 7 seconds on an average 5
  • 6. 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 P(+) P(+|+) P(+|+,+) P(+|-,+) P(+|+,+,+) P(+|-,+,+) P(+|+,-,+) P(+|+,+,-) P(+|+,+,+,+) P(+|-,+,+,+) P(+|+,-,+,+) P(+|+,+,-,+) P(+|+,+,+,-) P(+|-,-,+,+) P(+|-,+,-,+) P(+|-,+,+,-) all data 1st half 2nd half FIG. 3: Probability structure of the Sony bank USD/JPY rate. Each conditional probability has three bars. The upper one is the result of whole period 22/10/2001 to 31/05/2005. The middle is for the first half of the data. The lower is for the second half of the data. The error bars represent a 95% confidence interval on the mean and sometimes even shorter. This may be too frequent and small for practical use. However, the Sony bank rate essentially reflects mostly 0.1 yen price changes and about 20 minutes interval between price chages. Therefore, there is a better chance to take advantage of the Sony bank rate. In addition, we compare the probability structure of GE stock data on NYSE shown in [6]. The data was collected in December 2002. The number of GE data for one day is about 180,000, which is more frequent than the tick-by-tick data for USD/JPY. Figure 4 shows that the reliability of the probability structure increases as the sam- pling frequency increases, comparing tick-by-tick data for USD/JPY, the Sony bank USD/JPY rate and GE data on NYSE. The exceptions are regular “zigzag” motions such as P(+|−, +), P(+|+, −, +), P(+|−, +, −, +). In these cases, the probability structure of the Sony bank rate is more reliable than the tick-by-tick USD/JPY rate. 6
  • 7. 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 P(+) P(+|+) P(+|+,+) P(+|-,+) P(+|+,+,+) P(+|-,+,+) P(+|+,-,+) P(+|+,+,-) P(+|+,+,+,+) P(+|-,+,+,+) P(+|+,-,+,+) P(+|+,+,-,+) P(+|+,+,+,-) P(+|-,-,+,+) P(+|-,+,-,+) P(+|-,+,+,-) Sony bank rate tick-by-tick data GE FIG. 4: Comparison of the bias among the tick-by-tick data for GE on NYSE, tick-by-tick data for USD/JPY and Sony bank USD/JPY rates. The reliability of the probability structure increaces as the frequency of the data increaces. The error bars represent a 95% confidence interval on the mean. IV. CONCLUSION We analyzed the Sony bank USD/JPY rate, which closely reflects the market rate, with a focus on the up/down price movement, as opposed to the extent of price movement. The Sony bank has a lower sampling rate and shows larger price change deviations than the market rate. The main result is that a probability structure is observed in the Sony bank rate, in spite of the interval between data points of the Sony bank rate is about 20 minutes. However, this probability structure is not found in the data which has been sampled from tick-by-tick data at the same rate as the Sony bank rate. Therefore, the method of generating the Sony bank rate from the market rate can retain the probability structure as the sampling frequency decreases. There is a better chance to take advantage of the Sony bank rate in practical use. 7
  • 8. Acknowledgments I would like to thank Shigeru Ishi, President of the Sony bank, for kindly providing the Sony bank data. The NYSE data are from King’s College London. [1] M. M. Dacorogna, R.Gencay, U. A. Muller, R. B. Olsen, V. Pictet, in An Introduction to High-Frequency Finance (Academic Press, San Diego, 2001). [2] H. Takayasu (Ed), Empirical Science of Financial Fluctuations:The Advent of Econophysics. , (Springer-Verlag Tokyo, 2002). [3] T. Ohira, N. Sazuka, K. Marumo, T. Shimizu, M. Takayasu and H. Takayasu, Physica A, 308, N1-4, pp. 368–374, (2002). [4] N. Sazuka, T. Ohira, K. Marumo, T. Shimizu, M. Takayasu and H. Takayasu, Physica A, 324, N1-2, pp. 366–371, (2003). [5] Y. Zhang, Physica A 269, pp. 30–44, (1999). [6] N. Sazuka, Physica A, 355, pp. 183–189, (2005). 8