Matematiksel iktisat ve finans

Bir okuyucumuz su maili gondermis:

Bu modellerin hava tahminlerinde olduğu gibi kısa vade için işe yarıyabildiğini,uzun vadede ise isabet ihtimalinin düştüğünü biliyorum.Hava tahminlerinden daha komplexler çünkü insan faktörü-siyaset vs - daha fazla işin içine giriyor

Aracı kurumların ve bazı bireylerin borsa operasyonlarını sürekli gelişen matematiksel modelleme programlarıyla yaptığı ve bu işlemlerin toplamın %80 i gibi bir düzeye geldiği biliniyor.

Bu durum sadece teknik endikatörleri iyi kullanmasını bilen swing trader in sonu anlamına mı geliyor? Yoksa aracı kurumlara gitmeden de bu tür modelleri kişiselleştirip uygulama imkanımız var mı ?

Bu tip uygulamaların Türkiye deki aracı kurumlarda uygulanma durumu nedir?

Temel analiz ekolune eğilimli olduğunuzu biliyorum,ancak bu konuyla ilgili bir başlık açabilirsek teknik analizle ilgilenen çok sayıdaki okura yol gösterici olabilir diye düşünüyorum.

Bu soru bizim rdynk'in borsa modeli yazisinin uzerine iyi gitmis. Turkiye'deki traderlar ne kadar matematik modelleme kullaniyor bilemiyorum, ama kullandiklari konusunda simdiye kadar bir izlenimim olmadi. Belki bu konuda diger okuyucularimiz yardimci olabilir.

Ama ben size Amerika'da ne tur matematik modelleme kullanildigi konusunda soyle yardimci olabilirim. Mail kutuma gelen su mesaja bakarsaniz borsada oynamanin o kadar basit olmadigini, bu islerin "30000'de destek var, 35000'de kostek var" gibi yaklasimlarla yurumeyecegini gorursunuz. Amerika'da bu boyle ama. Turkiye'de dogru durust rekabet yok, araci kurumlarin tek amaci var, o da musterilerine mumkun oldugunca cok islem yaptirmak. Kimsenin getirileri maksimize etme, riskleri asgariye indirme gibi bir derdi yok. O yuzden ben yaptigim yarim yamalak analizlerle bile borsa getirisinin cok uzerinde getiriler elde edebiliyorum. O yuzden Amerika'da calisan bir modelci senede en azindan $250-$300 bin kazanirken Turkiye'de metelige kursun atiyor. Turkiye'de araci kurumlarin ve portfoy yoneticilerinin yapabilecegi o kadar cok sey var ki. Akilli 3-4 adam ciksa hem sektorde yonetilen para miktarini 10 katina cikarir hem de sektordeki araci kurum sayisini 10'da birine dusurur. (wink wink: Bende cok iyi fikirler var, ilgilenen girisimcilere duyurulur)

What is Insider Trading Anomaly
Recent Academic Studies on Insider Trading
Insider Trading in Netherlands
Insider Trading Returns
Definition of Insider Trading
Is Insider Trading Legal?
How Insiders Use Private Information and Don’t Get Caught?
SEC Regulation on Insider Trading: Section 10b

High-Frequency Finance and Quantitative Strategies
June 10-12, 2009
Courant Institute, Room 109, 251 Mercer Street New York, NY 10012

This 3-day workshop provides a thorough coverage of quantitative investment management and high frequency trading:
. Financial market microstructure for the practitioner
. Mechanics of trading (bkz. definition of insider trading)
. Common trading strategies
. How to work with high frequency data
. Estimation of transaction costs and market impact models
. Portfolio construction with the Black-Litterman model and robust optimization
. Portfolio optimization with transaction cost
. Optimal betting and execution strategies
. Simulation techniques
. Back-testing strategies
. Multi-period dynamic portfolio optimization with transaction costs
. Performance measurement
Dynamic programming, econometrics and model risk mitigation techniques are covered throughout the workshop.
Sessions are held from 8:30 a.m. to 5 p.m. over the three days at the Courant Institute of Mathematical Sciences, Room 109, 251 Mercer St., New York, New York. Continental breakfast and afternoon refreshments are provided.

Audience
Buy-side practitioners (portfolio managers and risk managers), sell-side practitioners (traders,financial engineers, quantitative analysts, research teams), and academics will deepen and broaden their understanding of the recipes they implement everyday and will learn the most cutting-edge techniques. Prerequisites for the workshop are undergraduate linear algebra, probability theory and some knowledge of mathematical finance at the level of a first term in an M.S. program. Some basic programming skills are a plus.

Day 1 - June 10, 2009
8:00-8:30 Registration and Opening Address
Petter Kolm, Deputy Director of Mathematics in Finance M.S. Program, CourantInstitute

Morning Session (8:30-12:30)
Part I. Markets and Data
Lee Maclin
. Mechanics of trading
o Limit and market orders
o The two-way double auction
o Order placement rules
o Exchange mechanisms
o Order management systems and trading platforms
o Direct market access pipes
. Working with high frequency data
o The TAQ data set
o The Nasdaq aggregate data set
Part II. Optimal Execution and Market Impact
Petter Kolm
. Introduction to the calculus of variations
. Optimal execution of portfolio transactions
o Permanent and temporary impact
o Implementation shortfall
o The Almgren-Chriss model
o The efficient frontier of optimal execution
. Market impact models
o Almgren et al.
o Models utilizing public data
Afternoon Session (2:00-5:00)
Part III. Robust Portfolio Optimization with Transaction Costs
Petter Kolm
. The Black-Litterman model
. Robust portfolio optimization
. Incorporating transaction costs into portfolio optimization
Part IV. Risk Models
Eran Fishler
. Factor models and covariance matrix estimation
o Multifactor models
o Some results from random matrix theory
o Covariance matrix "cleaning"
o PCA, linear regression, shrinkage and ridge regression
o Shrinkage estimators of regression coefficients
o Ridge regression

Day 2 - June 11, 2009
Morning Session (8:30-12:30)

Part I. Factor-Based Quantitative Trading Strategies
Joseph Cerniglia
. Quantitative trading strategies
. Standard themes
. Data
. Back-testing methodologies
II. Time-Series Analysis
Farshid Asl
. ARIMA models
. Forecasting
Afternoon Session (2:00-5:00)
Part III. Trading with Time-Series Models
Farshid Asl
Part IV. Statistical Arbitrage in the U.S. Equities Market
Marco Avellaneda
. Long-short equity
. Hedging instruments
. Back-testing

Day 3 - June 12, 2009
Morning Session (8:30-12:30)

Part I. The Optimal Theta Framework
Lee Maclin
. Introduction: Dynamic portfolio analysis
. Traditional rebalancing
. The dynamic portfolio framework
. A simulation framework for dynamic portfolio analysis
. Optimal betting (Optimal f) and optimal growth portfolios
Part II. Combining Portfolio
Optimization and Optimal Execution
Petter Kolm
. The Engle-Ferstenberg model
. Sharpe ratio with market impact
. Dynamic programming
Afternoon Session (2:00-5:00)
Part III. Control Theory
Farshid Asl
. Introduction
. Filtering
. State space models
Part IV. Dynamic Portfolio Analysis
Petter Kolm
. Stochastic Linear Quadratic Gaussian (LQG) Regulator
. Dynamic portfolio analysis with transaction costs