Day Trading Momentum Stocks

Day trading momentum stocks involves taking advantage of rapid price movements within a single trading day. 

Momentum is a key variable that many traders want part of their process. 

If you look at most price charts, they might look semi-random, but you always notice that they tend to move in trends a lot too.

Institutional traders and investors have developed sophisticated methods to measure momentum and use ensemble techniques to improve their trading strategies.

We’ll take a look at the type of things that they do.

Key Takeaways – Day Trading Momentum Stocks

• Momentum trading is where traders take advantage of the continuing trend of an asset’s price – buying when the price is rising and selling when it starts to decline.
• Popular indicators – Common momentum indicators in trading include the Relative Strength Index (RSI), Moving Average Convergence Divergence (MACD), and the Stochastic Oscillator, which help traders evaluate the speed and strength of price changes.
• Combine models – Use multiple indicators, frameworks, and analysis methods to confirm momentum signals.
• Process
  • Identify Indicators – Select multiple momentum indicators (e.g., price momentum, earnings momentum, relative strength, custom-made).
  • Backtest – Test the indicators historically, individually, and in combination.
  • Optimize – Refine indicator parameters and weightings to maximize performance.
  • Develop Rules – Establish clear entry and exit criteria based on the optimized indicators.
  • Implement – Trade systematically or use the rules to guide discretionary decisions.

Understanding Momentum Trading

What Is Momentum Trading?

Momentum trading is a strategy that tries to profit from the continuation of existing trends in the market. 

Traders buy securities that are trending upwards and sell those that are trending downwards.

They’re essentially betting that these trends will persist for a certain period.

Importance in Day Trading

In day trading, momentum is often used because traders rely on quick price movements to make profits. 

Traditional Momentum Indicators

Moving Averages

Moving averages smooth out price data to identify trends.

Simple moving averages (SMA) and exponential moving averages (EMA) are commonly used to assess momentum by comparing short-term averages to long-term averages.

Relative Strength Index (RSI)

The RSI measures the speed and change of price movements, oscillating between zero and 100.

An RSI above 70 typically indicates overbought conditions, while below 30 suggests oversold conditions.

Of course, 70/30 is arbitrary. It could be 75/25, 80/20, or something else.

In terms of more advanced momentum measuring techniques…

Statistical Models

Institutions use statistical models like the Autoregressive Integrated Moving Average (ARIMA) to forecast future price movements based on past data.

These models account for trends, seasonality, and cyclicality in price data.

Machine Learning Approaches

Machine learning algorithms analyze large datasets to identify patterns not visible through traditional analysis. 

Techniques like Support Vector Machines (SVM) and Neural Networks help in predicting momentum by learning from historical price movements and other relevant factors.

Ensemble Methods in Momentum Trading

Ensemble methods combine multiple models to improve prediction accuracy, a technique increasingly adopted by institutional traders.

The idea is that combining diverse models can compensate for individual weaknesses.

For example, some machine learning models are not good at statistical accuracy, but combining them with purely statistical models can take advantage of what they’re good at while protecting them from what they’re bad at.

Definition of Ensemble Methods

Ensemble methods involve integrating several predictive models to produce a single, more accurate forecast. 

Application in Trading Strategies

By aggregating signals from different models, traders can make more informed decisions. 

Ensemble methods help in filtering out noise and reducing the impact of outliers on trading signals.

Advantages Over Single Models

Ensemble methods offer improved accuracy and robustness. 

They reduce the risk associated with relying on a single model that might fail under certain market environments.

Examples of Ensemble Methods

Ensemble Method 1: Weighted Average Crossover

• Indicator = Moving Average Crossover
• Inputs:
  • 5-day Exponential Moving Average (EMA)
  • 20-day Simple Moving Average (SMA)
  • 50-day EMA
• Weightings:
  • 5-day EMA: 50%
  • 20-day SMA: 30%
  • 50-day EMA: 20%

Trading Logic:

• A buy signal is generated when the 5-day EMA crosses above the 20-day SMA and the 50-day EMA.
  • The strength of the signal is determined by the weighted average of each individual crossover signal (e.g., 1 if 5-day EMA > 20-day SMA, 0 if below).
• A sell signal is generated when the 5-day EMA crosses below the 20-day SMA and the 50-day EMA.

Ensemble Method 2: Dual Momentum with Volatility Filter

• Indicator = Combined Momentum Score
• Inputs:
  • Price Momentum = 12-month return
  • Relative Momentum = 12-month return relative to a benchmark index (e.g., S&P 500)
  • Volatility Filter = 20-day Average True Range (ATR)
• Weightings:
  • Price Momentum = 60%
  • Relative Momentum = 40%

Trading Logic:

• Calculate a momentum score by combining the weighted price momentum and relative momentum.
• Enter a long position if the momentum score is positive and the 20-day ATR is below a predefined threshold (to filter out volatile periods).
• Exit the long position if the momentum score turns negative or the 20-day ATR crosses above the threshold.

How Institutional Traders Use Ensemble Methods

Institutions leverage ensemble methods to enhance their momentum trading strategies in various ways.

Combining Multiple Models

Institutional traders might combine fundamental analysis models, technical indicators, and machine learning algorithms. 

For example, they could integrate signals from a momentum-based model with a mean-reversion model to capture different market behaviors.

Risk Management

Considering multiple perspectives can help traders better understand the probability of different outcomes.

All models are simplifications of reality.

With different processing of data or inputs, they will each see things differently.

Real-Time Data Processing

Institutions use high-frequency data and real-time processing to feed their ensemble models. 

This capability allows them to react quickly to market changes, a key factor in day trading.

Overall Process

Identify Indicators

Select multiple momentum indicators like price momentum, earnings momentum, relative strength, or custom-made indicators.

Backtest

Test these indicators historically, both individually and in combination, to determine their effectiveness in predicting price movements.

What shows good results on past data?

Consider synthetic data that emulates real-world markets across various scenarios.

If nothing is good, then it’s back to step 1 (idea generation).

Optimize

Refine the parameters and weightings of the indicators to maximize trading performance.

For example, if you’re using a mix (ensemble), you might tweak the weights.

This helps in identifying the best configuration for profitable trades.

Develop Rules

Create clear and specific entry and exit rules based on the optimized indicators.

These rules define when to buy and sell based on momentum signals.

Implement

Trade systematically using automated systems or apply the developed rules to guide discretionary decisions.

Consistency in following these rules gives discipline to your trading.

Let’s go through some ways of measuring momentum.

To have unique perspectives on what momentum is and trading it, you’ll need to:

• Use more than one indicator
• Have unique interpretations of existing ones
• Create your own

1. Moving Average Convergence Divergence (MACD)

The MACD is a momentum oscillator that shows the relationship between two moving averages of a stock’s price.

How to Calculate:

• MACD Line – Subtract the 26-period Exponential Moving Average (EMA) from the 12-period EMA.
• Signal Line – Calculate the 9-period EMA of the MACD Line.
• Histogram – The difference between the MACD Line and the Signal Line.

Usage:

• Bullish Signal – When the MACD Line crosses above the Signal Line.
• Bearish Signal – When the MACD Line crosses below the Signal Line.

Example – If the 12-period EMA is $30 and the 26-period EMA is $28:

• MACD Line: $30 – $28 = $2
• If the 9-period EMA of the MACD Line is $1.5, then the Histogram is $2 – $1.5 = $0.5

2. Relative Strength Index (RSI)

RSI measures the speed and change of price movements, oscillating between 0 and 100.

How to Calculate:

• RSI = 100 – [100 / (1 + RS)]
• RS (Relative Strength) – Average Gain over N periods / Average Loss over N periods.

Usage:

• Overbought Condition – RSI above 70 suggests a potential downturn.
• Oversold Condition – RSI below 30 indicates a possible upturn.

Example – Over 14 periods, if the average gain is $0.8 and the average loss is $0.2:

• RS = $0.8 / $0.2 = 4
• RSI = 100 – [100 / (1 + 4)] = 100 – [100 / 5] = 80

3. Rate of Change (ROC)

ROC measures the percentage change between the current price and the price a certain number of periods ago.

How to Calculate:

• ROC = [(Current Price – Price N periods ago) / Price N periods ago] * 100

Usage:

• Positive ROC = Indicates upward momentum.
• Negative ROC = Suggests downward momentum.

Example – If the current price is $55 and the price 10 periods ago was $50:

• ROC = [($55 – $50) / $50] * 100 = 10%

4. Stochastic Oscillator

This indicator compares a particular closing price to a range of its prices over a certain period.

How to Calculate:

• %K = [(Current Close – Lowest Low) / (Highest High – Lowest Low)] * 100
• %D: 3-period moving average of %K.

Usage:

• Overbought = Above 80
• Oversold = Below 20
• 80/20 is arbitrary and it could be some other thresholds (e.g., 85/15, 90/10, 75/25, etc.)

Example – If the current close is $70, the lowest low over 14 periods is $60, and the highest high is $80:

• %K = [($70 – $60) / ($80 – $60)] * 100 = 50%

5. On-Balance Volume (OBV)

OBV measures buying and selling pressure as a cumulative indicator.

How to Calculate:

• OBV – Add volume on up days and subtract volume on down days.

Usage:

• Rising OBV – Confirms an uptrend.
• Falling OBV – Confirms a downtrend.

Example:

• Day 1: Close up, Volume = 1 million shares, OBV increases by 1 million.
• Day 2: Close down, Volume = 500,000 shares, OBV decreases by 500,000.

6. Average Directional Index (ADX)

ADX quantifies the strength of a trend.

How to Calculate:

• ADX – Based on the moving averages of expanding price range values.

Usage:

• Strong Trend – ADX above 25.
• Weak Trend – ADX below 20.

Example – If the calculated ADX is 30, it suggests a strong trend is present.

7. Volume-Weighted Average Price (VWAP)

VWAP gives the average price a security has traded at throughout the day, based on both volume and price.

How to Calculate:

• VWAP = Cumulative (Price * Volume) / Cumulative Volume

Usage:

• Price Above VWAP – Indicates bullish momentum.
• Price Below VWAP – Indicates bearish momentum.

Example – If during the day, cumulative price-volume is $5 million and cumulative volume is 100,000 shares:

• VWAP = $5,000,000 / 100,000 = $50

8. Momentum Indicator

Measures the rate of change of a stock’s price.

How to Calculate:

• Momentum = [Current Price – Price N periods ago] / (Some Unit of Time)

Usage:

• Positive Value – Upward momentum.
• Negative Value – Downward momentum.

Example – Current price is $105, and price 10 days ago was $100:

• Momentum = ($105 – $100) / 10 = $0.50 per day

9. Bollinger Bands

These are volatility bands placed above and below a moving average.

How to Calculate:

• Middle Band – 20-day simple moving average (SMA).
• Upper Band: Middle Band + (2 x standard deviation).
• Lower Band: Middle Band – (2 x standard deviation).

Usage:

• Price Touching Upper Band – May indicate overbought conditions.
• Price Touching Lower Band – Could indicate oversold conditions.

Example – If the 20-day SMA is $50 and the standard deviation is $2:

• Upper Band: $50 + (2 x $2) = $54
• Lower Band: $50 – (2 x $2) = $46

10. Ichimoku Cloud

Provides support and resistance levels, trend direction, and momentum.

How to Calculate:

• Conversion Line (Tenkan-sen): (9-period high + 9-period low) / 2
• Base Line (Kijun-sen): (26-period high + 26-period low) / 2
• Leading Span A (Senkou Span A): (Conversion Line + Base Line) / 2, plotted 26 periods ahead.
• Leading Span B (Senkou Span B): (52-period high + 52-period low) / 2, plotted 26 periods ahead.

Usage:

• Price Above Cloud: Bullish signal.
• Price Below Cloud: Bearish signal.

Example – If the 9-period high is $60 and the 9-period low is $50:

• Conversion Line: ($60 + $50) / 2 = $55

11. Commodity Channel Index (CCI)

Measures the difference between a security’s price change and its average price change.

How to Calculate:

• CCI = (Typical Price – SMA) / (0.015 x Mean Deviation)
• Typical Price = (High + Low + Close) / 3

Usage:

• CCI Above +100 = Overbought condition.
• CCI Below -100 = Oversold condition.

Example – If the typical price is $70, SMA is $65, and mean deviation is $2:

• CCI = ($70 – $65) / (0.015 x $2) = $5 / $0.03 = 166.67

12. Statistical Models (ARIMA)

Autoregressive Integrated Moving Average models forecast future price movements based on past data.

How to Use:

• Fit an ARIMA model to historical price data and use the model to forecast future prices.

Example – An ARIMA(1,1,1) model predicts tomorrow’s price based on today’s price and the error from previous predictions.

Note that ARIMA models have been used in trying to predict price movements in markets since the 1970s and their validity on forward markets (in isolation) might not be a strategy in itself.

It’s just one analytical method among many.

13. Machine Learning Models

Algorithms like Support Vector Machines (SVM) or Neural Networks predict price movements by learning from historical data.

How to Use:

• Collect historical price and volume data.
• Train the model to recognize patterns associated with momentum.
• Use the trained model to predict future momentum.

Example – A neural network predicts a 5% increase in stock price based on input features like RSI, MACD, and volume trends based on past movements. Also be sure not to overly optimize machine learning models based on the past because markets change over time

14. Ensemble Methods

Combine multiple models to improve prediction accuracy.

How to Use:

• Aggregate signals from different indicators and models.
• For decision-making, use…
  • Majority voting (all equal-weighted)
  • Weighted averages

Example – If three models predict:

• Model A = Buy
• Model B = Buy
• Model C = Sell
• The ensemble decision is to Buy based on majority.

An ensemble might weight them separately based on their confidence in their predictive capacity.

For example, if you weight Model A at 20%, Model B at 25%, and Model C at 55%, the above model might actually flip to a Sell.

15. Alternative Data Analysis

Use non-traditional data sources to gauge momentum.

How to Use:

• Social Media Sentiment – Analyze Twitter/X or Reddit posts for sentiment trends. Naturally, this will be skewed more toward retail sentiment.
• News Feeds – Use natural language processing to assess the tone of news articles.
• Web Traffic – Monitor increases in website visits for companies.

Example – An uptick in positive social media mentions correlates with upward stock movement, indicating positive momentum.

16. High-Frequency Trading Signals

Use algorithms to identify and act on momentum at very short time intervals.

This is technically difficult to do.

How to Use:

• Use automated trading systems that execute trades in milliseconds.
• Understand concepts like book skew.
• Use tick data to analyze momentum shifts.

Example – An algorithm detects a sudden increase in bid prices and executes a buy order before the price rises further.

• Related: HFT Strategies

17. Pair Trading

Identify two correlated stocks and trade based on the divergence from their historical correlation.

How to Use:

• Calculate Spread – Difference in prices between the two stocks. This can be done in various ways, such as P/E ratio, DCF analysis, book values, or proprietary valuation methods.

Example – If Stock A and Stock B usually trade at similar prices but diverge holding fundamental changes constant, buy the underperforming stock and sell the outperforming one, expecting them to revert.

18. Sector Rotation Strategies

Measure momentum by analyzing performance across different sectors.

How to Use:

• Compare sector indices to identify which sectors are gaining momentum.
• Rotate investments into high-momentum sectors.
• Some with a market-neutral focus will short the low-momentum sectors.

Example – If the technology sector shows increasing returns compared to others, shift capital into tech stocks.

19. Event-Driven Momentum

Use corporate events to predict momentum shifts.

How to Use:

• Earnings Reports – Positive earnings surprises often lead to upward momentum. Of course, underperforming can lead to price falls.
• Mergers and Acquisitions – Announcements can cause large price movements.

Example – A company reports earnings that exceed expectations, leading to increased buying and upward momentum.

20. Behavioral Indicators

Analyze behavior patterns to predict momentum.

How to Use:

• Put/Call Ratios – High ratios may indicate bearish sentiment.
• Short Interest – High short interest could lead to a short squeeze and upward momentum.
• Volatility Index (VIX) – High readings suggest fear and potential market downturns. Low readings may indicate complacency.
• Investor Sentiment Surveys – Excessively bullish readings can be contrarian and signal a market top. Extremely bearish readings may indicate a bottom.
• Margin Debt – Rising levels can fuel market advances but also increase risk, while declining levels may signal decreased confidence.
• Market Breadth – Strong breadth (more advancing than declining stocks) indicates a healthy market, while weak breadth suggests underlying weakness.
• Advance-Decline Line – A rising line confirms an uptrend. A falling line signals underlying weakness.
• New Highs/New Lows – Increasing new highs indicate strong momentum, while increasing new lows suggest broad market weakness.
• Fund Flows – Strong inflows into equity funds suggest positive sentiment, while outflows may indicate caution.
• Insider Trading – Increased insider buying can be a bullish signal, while increased insider selling may suggest caution.

Example – An unusually high short interest in a stock may set the stage for a rapid price increase if positive news triggers short covering.