TOPIC 5
TRADING STRATEGIES
Checklist for Making a Trade
 What is the direction of the overall market?
 What is the direction of the market sector?
 What are major, intermediate and minor trends?
 Are there any major reversal patterns visible?
 Are there any major continuation pattern visible?
 What are the markings on candlesticks charts?
 What are the price objectives in those patterns?
 What are important support and resistance levels?
 What are the anticipated breakout or retracement levels?
 Which way the moving averages point?
 Are oscillators overbought/oversold?
 Are there any divergences visible on oscillators?
 Does volume confirm the trend? 2
Ask These Questions before Entering
into Trade
 Which way will the market/ stock trend?
 Am I going to buy or sell?
 How much am I willing to lose if I’m wrong?
 How many units?
 What is my entry point?
 What is my price objective?
 Where will I keep the stop loss?
 What type of order will I use?
3
 Price forecast
 Identify which way the market is expected to move
 Decide on long/short position
 Based on technical or fundamentals trading rules
 Trading tactics
 Decide on timing of entry/exit point
 Decide on type of the order
 Money management
 Allocate funds to a particular trade
 Decide to be aggressive or conservative
 Decide the risk to reward and use of stop loss
4
Elements of Successful Trading Strategy
Testing of Trading Strategy
 A trading strategy is a tested set of rules selected
to do trading
 An edge in trading is an exploitable statistical
advantage based on market behaviour that is
likely to recur in the future
 Over different markets
 Over varies market conditions
 How do you know that your trading strategy or
trading rule has an edge?
 How to develop a workable trading strategy?
5
Back-Testing
 The simulation of a trading strategy
 Back-testing implies that the forecasting models
are fitted and tested using historical data.
Usually, the entire available data set is split into
 In-sample data – the earlier data is used for the
calibration of the predicting model
 Out-of-sample testing - the other is reserved for
testing of the calibrated model.
 In simple models, it is usually assumed that the
testing sample is variance-stationary; that is, the
sample volatility is constant
6
Testing Process
 Rules
 Long when 50D EMA cuts above the 100D EMA
 Variables
 Shorter EMA and longer EMA
 Parameters
 50D and 100D
 Filters
 Trade when RSI is over 50 and no trade when RSI is over 70
 Data sample
 From 2008 to 2018
 Outputs are the trading signals
 Analyze results
 About Profits – winning %, profit factor, P&L
 About Risks – Sharpe, Sortino, max drawdown
 Optimize parameters
7
Weakness of Back-Testing
 Curve overfitting
 The model fits closely or exactly to a particular set of
data (historical) but fails to fit new data (future)
 too good to be true
 Extrapolation
 The result from in-sample prediction may not be
generalized to out-of-sample data
 Therefore, periodic adjustment or updates of the
strategy are required
8
Fundamental Factors Back-Testing
 Select a factor
 Any common characteristic that is important in
explaining the risk and return of a group of securities
 Value (high P/E vs. low P/E; high P/B vs. low P/B or
high dividend yield vs. low dividend yield)
 Momentum (outperforming vs. underperforming)
 Quality (high accruals vs. low accruals)
9
How to Back-Test a Factor
 We want to test a strategy of buying cheap stocks (low
P/E) and avoiding expensive stocks(high P/E)
 Stocks are sorted into 5 Quintiles (buckets) with the
cheapest in Quintile 1
 Long stocks in Q1 and/or short stocks in Q2 and
rebalanced periodically.
10
A universe of 1,000 stocks
Q1 Q2 Q3 Q4 Q5
Factor Back-Testing Steps
1. Define an investment universe
2. Identify and describe the factor to be tested.
3. Describe the investment strategy that
underpins your factor thesis
4. Grouping methodology (Deciles, Quartile)
5. Determine your holding period or rebalancing
assumption
6. Check for data availability
7. Look at the metrics that evaluate the
performance of the factor trading strategy
11
Other Examples of Factors
 Profitability
 ROE, ROA, NI, earnings surprise, EPS
 Volatility
 New 52 week high,
 Growth
 Annual earnings grow over 5 years, quarterly
earnings growth, past 12 months return
 Leverage
 Debt ratio
 Institutional ownership
 Volume
 Corporate governance 12
FTST
13
14
FTST
Technical Rules Back-Testing
 Determine the trading signal (moving average
cross-over)
 Decide how you trade on the signals, when to
entre and exit, how frequently you review them
and rebalance your positions
15
BT
16
BT
17
 practical performance measures should be evaluated after
fees.
 Percent of Winning Trades
 Ratio of average winning amount to average losing amount
 Profit factor is ratio of net profit versus net loss
 Hit rate is the percentage of outperformers
 The Sharpe ratio is often used in performance analysis.
 Sometimes, the Sortino ratio is chosen instead.
 Information ratio is used if a trading strategy performance
is compared with performance of an index
 The maximum drawdown (MD) is another important risk
measure
 MD is the largest drop of price after its peak.
18
Performance Measures
For Continued Success in Use of TA to
Build Trading Strategies
 There is no hard and fast rule to determine the
input parameters
 diversification across different trading strategies
and/or instruments.
 Use a combination of strategies
 In a diversified portfolio
 TA can be used in combination of fundamental value
modeling techniques.