Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading...
Transcript of Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading...
![Page 1: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/1.jpg)
Introduction to Algorithmic Trading StrategiesLecture 6
Technical Analysis: Linear Trading Rules
Haksun [email protected]
www.numericalmethod.com
![Page 2: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/2.jpg)
Outline Moving average crossover The generalized linear trading rule P&Ls for different returns generating processes Time series modeling
![Page 3: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/3.jpg)
References Emmanual Acar, Stephen Satchell. Chapters 4, 5 & 6, Advanced Trading Rules, Second Edition. Butterworth‐Heinemann; 2nd edition. June 19, 2002.
![Page 4: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/4.jpg)
Assumptions of Technical Analysis History repeats itself. Patterns exist.
![Page 5: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/5.jpg)
Does MA Make Money? Brock, Lakonishok and LeBaron (1992) find that a subclass of the moving‐average rule does produce statistically significant average returns in US equities.
Levich and Thomas (1993) find that a subclass of the moving‐average rule does produce statistically significant average returns in FX.
![Page 6: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/6.jpg)
Moving Average Crossover Two moving averages: slow ( ) and fast ( ). Monitor the crossovers.
,
Long when . Short when .
![Page 7: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/7.jpg)
How to Choose and ? It is an art, not a science (so far). They should be related to the length of market cycles. Different assets have different and . Popular choices: (150, 1) (200, 1)
![Page 8: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/8.jpg)
AMA(n , 1)
iff
iff
![Page 9: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/9.jpg)
GMA(n , 1)
iff
∑ (by taking log)
iff
∑ (by taking log)
![Page 10: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/10.jpg)
What is ?
![Page 11: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/11.jpg)
Acar Framework Acar (1993): to investigate the probability distribution of realized returns from a trading rule, we need the explicit specification of the trading rule the underlying stochastic process for asset returns the particular return concept involved
![Page 12: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/12.jpg)
Empirical Properties of Financial Time Series Asymmetry Fat tails
![Page 13: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/13.jpg)
Knight‐Satchell‐Tran Intuition Stock returns staying going up (down) depends on the realizations of positive (negative) shocks the persistence of these shocks
Shocks are modeled by gamma processes. Persistence is modeled by a Markov switching process.
![Page 14: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/14.jpg)
Knight‐Satchell‐Tran Process
: long term mean of returns, e.g., 0 , : positive and negative shocks, non‐negative, i.i.d
![Page 15: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/15.jpg)
Knight‐Satchell‐Tran
Zt = 0 Zt = 1q p
1‐q
1‐p
![Page 16: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/16.jpg)
Stationary State
, with probability , with probability
![Page 17: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/17.jpg)
GMA(2, 1) Assume the long term mean is 0, .
![Page 18: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/18.jpg)
Naïve MA Trading Rule Buy when the asset return in the present period is positive.
Sell when the asset return in the present period is negative.
![Page 19: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/19.jpg)
Naïve MA Conditions The expected value of the positive shocks to asset return >> the expected value of negative shocks.
The positive shocks persistency >> that of negative shocks.
![Page 20: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/20.jpg)
Period Returns
Sell at this time point
0
0 1
hold
![Page 21: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/21.jpg)
Holding Time Distribution
![Page 22: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/22.jpg)
Conditional Returns Distribution (1)
|∑
∑
∑
⋯
![Page 23: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/23.jpg)
Unconditional Returns Distribution (2)
∑
![Page 24: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/24.jpg)
Long‐Only Returns Distribution
Proof: make
![Page 25: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/25.jpg)
I.I.D Returns Distribution
Proof: 1 make Π 1
![Page 26: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/26.jpg)
Expected Returns
When is the expected return positive? , shock impact
≫ , shock impact Π 1 , if , persistence
![Page 27: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/27.jpg)
GMA(∞,1) Rule
![Page 28: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/28.jpg)
GMA(∞,1) Returns Process
![Page 29: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/29.jpg)
Returns As a MA(1) Process
![Page 30: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/30.jpg)
GMA(∞,1) Expected Returns
![Page 31: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/31.jpg)
MA Using the Whole History An investor will always expect to lose money using GMA(∞,1)!
An investor loses the least amount of money when the return process is a random walk.
![Page 32: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/32.jpg)
Optimal MA Parameters So, what are the optimal and ?
![Page 33: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/33.jpg)
Linear Technical Indicators As we shall see, a number of linear technical indicators, including the Moving Average Crossover, are really the “same” generalized indicator using different parameters.
![Page 34: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/34.jpg)
The Generalized Linear Trading Rule A linear predictor of weighted lagged returns ∑
The trading rule Long: 1, iff, 0 Short: 1, iff, 0
(Unrealized) rule returns
if 1 if 1
![Page 35: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/35.jpg)
Buy And Hold
![Page 36: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/36.jpg)
Predictor Properties Linear Autoregressive Gaussian, assuming is Gaussian If the underlying returns process is linear, yields the best forecasts in the mean squared error sense.
![Page 37: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/37.jpg)
Returns Variance
![Page 38: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/38.jpg)
Maximization Objective Variance of returns is inversely proportional to expected returns.
The more profitable the trading rule is, the less risky this will be if risk is measured by volatility of the portfolio.
Maximizing returns will also maximize returns per unit of risk.
![Page 39: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/39.jpg)
Expected Returns
![Page 40: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/40.jpg)
Truncated Bivariate Moments Johnston and Kotz, 1972, p.116
Correlation: Corr ,
![Page 41: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/41.jpg)
Expected Returns As a Weighted Sum
a term for volatility a term for drift
![Page 42: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/42.jpg)
Praetz model, 1976
, the frequency of short positions
![Page 43: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/43.jpg)
Comparison with Praetz model
the probability of being short
increased variance
Random walk implies .
![Page 44: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/44.jpg)
Biased Forecast A biased (Gaussian) forecast may be suboptimal. Assume underlying mean . Assume forecast mean .
![Page 45: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/45.jpg)
Maximizing Returns Maximizing the correlation between forecast and one‐ahead return.
First order condition:
![Page 46: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/46.jpg)
First Order Condition
0
![Page 47: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/47.jpg)
Fitting vs. Prediction If process is Gaussian, no linear trading rule obtained from a finite history of can generate expected returns over and above .
Minimizing mean squared error maximizing P&L. In general, the relationship between MSE and P&L is highly non‐linear (Acar 1993).
![Page 48: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/48.jpg)
Technical Analysis Use a finite set of historical prices. Aim to maximize profit rather than to minimize mean squared error.
Claim to be able to capture complex non‐linearity. Certain rules are ill‐defined.
![Page 49: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/49.jpg)
Technical Linear Indicators For any technical indicator that generates signals from a finite linear combination of past prices Sell: 1 iff ∑ 0
There exists an (almost) equivalent AR rule. Sell: 1 iff δ ∑ 0
ln
∑ , ∑
![Page 50: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/50.jpg)
Conversion Assumption
Monte Carlo simulation: 97% accurate 3% error.
![Page 51: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/51.jpg)
Example Linear Technical Indicators Simple order Simple MA Weighted MA Exponential MA Momentum Double orders Double MA
![Page 52: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/52.jpg)
Returns: Random Walk With Drift
The bigger the order, the better. Momentum > SMAV > WMAV
How to estimate the future drift? Crystal ball? Delphic oracle?
![Page 53: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/53.jpg)
Results
![Page 54: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/54.jpg)
Results
![Page 55: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/55.jpg)
Returns: AR(1)
Auto‐correlation is required to be profitable. The smaller the order, the better. (quicker response)
![Page 56: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/56.jpg)
Results
![Page 57: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/57.jpg)
ARMA(1, 1)
Prices tend to move in one direction (trend) for a period of time and then change in a random and unpredictable fashion. Mean duration of trends:
Information has impacts on the returns in different days (lags). Returns correlation:
AR MA
![Page 58: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/58.jpg)
Results
no systematic winner
optimal order
![Page 59: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/59.jpg)
ARIMA(0, d, 0)
Irregular, erratic, aperiodic cycles.
![Page 60: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/60.jpg)
Results
![Page 61: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/61.jpg)
ARCH(p)
are the residuals When , .
residual coefficients as a function of lagged squared
residuals
![Page 62: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/62.jpg)
AR(2) – GARCH(1,1)AR(2) GARCH(1,1)
innovations
ARCH(1): lagged squared residuals
lagged variance
![Page 63: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/63.jpg)
Results The presence of conditional heteroskedasticity will not drastically affect returns generated by linear rules.
The presence of conditional heteroskedasticity, if unrelated to serial dependencies, may be neither a source of profits nor losses for linear rules.
![Page 64: Introduction to Algorithmic Trading Strategies Lecture 6Introduction to Algorithmic Trading Strategies Lecture 6 Technical Analysis: Linear Trading Rules Haksun Li haksun.li@numericalmethod.com](https://reader030.fdocuments.in/reader030/viewer/2022040107/5ed81b5fd9463e16115a23dd/html5/thumbnails/64.jpg)
Conclusions Trend following model requires positive (negative) autocorrelation to be profitable. What do you do when there is zero autocorrelation?
Trend following models are profitable when there are drifts. How to estimate drifts?
It seems quicker response rules tend to work better. Weights should be given to the more recent data.