Quantitative Models and Construction Methods Barra Style Factors - **Model Name**: Barra Style Factors - **Construction Idea**: The model tracks the performance of common style factors such as size, value, growth, and profitability in the market[3][13] - **Specific Construction Process**: The factors are constructed based on predefined dimensions: - Size factor: Measures the impact of market capitalization - Book-to-market ratio factor: Captures the value dimension - Growth factor: Reflects growth characteristics - Profitability factor: Tracks expected earnings performance[3][13] - **Evaluation**: The factors provide insights into the performance of different market styles, helping to understand market trends and dynamics[3][13] Open-Source Trading Behavior Factors - **Factor Name**: Ideal Reversal Factor - **Construction Idea**: Identifies the strongest reversal days by analyzing the average transaction size of large trades[4][14] - **Specific Construction Process**: 1. Retrieve the past 20 trading days' data for a stock 2. Calculate the average transaction size (transaction amount/number of transactions) for each day 3. Identify the 10 days with the highest transaction sizes and sum their returns (M_high) 4. Identify the 10 days with the lowest transaction sizes and sum their returns (M_low) 5. Compute the factor as \( M = M_{high} - M_{low} \)[43] - **Evaluation**: Captures micro-level reversal forces in the market, providing a unique perspective on trading behavior[4][14] - **Factor Name**: Smart Money Factor - **Construction Idea**: Tracks institutional trading activity by analyzing minute-level price and volume data[4][14] - **Specific Construction Process**: 1. Retrieve the past 10 days' minute-level data for a stock 2. Construct the indicator \( S_t = \frac{|R_t|}{V_t^{0.25}} \), where \( R_t \) is the return at minute \( t \) and \( V_t \) is the volume at minute \( t \) 3. Sort minute-level data by \( S_t \) in descending order and select the top 20% of minutes by cumulative volume as "smart money" trades 4. Calculate the volume-weighted average price (VWAP) for smart money trades (\( VWAP_{smart} \)) and all trades (\( VWAP_{all} \)) 5. Compute the factor as \( Q = \frac{VWAP_{smart}}{VWAP_{all}} \)[42][44] - **Evaluation**: Effectively identifies institutional trading patterns, offering a valuable alpha source[4][14] - **Factor Name**: APM Factor - **Construction Idea**: Measures the difference in stock behavior between morning (or overnight) and afternoon trading sessions[4][14] - **Specific Construction Process**: 1. Retrieve the past 20 days' data for a stock 2. Calculate daily overnight and afternoon returns for both the stock and the market index 3. Perform a regression of stock returns on market index returns to obtain residuals 4. Compute the difference between overnight and afternoon residuals for each day 5. Calculate the statistic \( \text{stat} = \frac{\mu(\delta_t)}{\sigma(\delta_t)/\sqrt{N}} \), where \( \mu \) is the mean, \( \sigma \) is the standard deviation, and \( N \) is the number of observations 6. Regress the statistic on momentum factors and use the residual as the APM factor[43][45][46] - **Evaluation**: Highlights intraday trading behavior differences, providing insights into market dynamics[4][14] - **Factor Name**: Ideal Amplitude Factor - **Construction Idea**: Measures the structural differences in amplitude information between high and low price states[4][14] - **Specific Construction Process**: 1. Retrieve the past 20 trading days' data for a stock 2. Calculate the daily amplitude as \( \text{Amplitude} = \text{(High Price/Low Price)} - 1 \) 3. Compute the average amplitude for the top 25% of days by closing price (V_high) 4. Compute the average amplitude for the bottom 25% of days by closing price (V_low) 5. Compute the factor as \( V = V_{high} - V_{low} \)[48] - **Evaluation**: Captures structural differences in price amplitude, offering a unique perspective on market behavior[4][14] - **Factor Name**: Composite Trading Behavior Factor - **Construction Idea**: Combines the above trading behavior factors using ICIR-based weights to enhance performance[32] - **Specific Construction Process**: 1. Perform industry-level outlier removal and standardization for each factor 2. Use the past 12 months' ICIR values as weights to combine the factors 3. Construct the composite factor as a weighted sum of the individual factors[32] - **Evaluation**: Demonstrates superior performance in small and mid-cap stock pools, providing robust alpha generation[32] --- Backtesting Results of Models and Factors Barra Style Factors - **Size Factor**: Return of -0.18% in November 2025[3][13] - **Book-to-Market Ratio Factor**: Return of 0.20% in November 2025[3][13] - **Growth Factor**: Return of -0.23% in November 2025[3][13] - **Profitability Factor**: Return of -0.35% in November 2025[3][13] Open-Source Trading Behavior Factors - **Ideal Reversal Factor**: - IC: -0.049 - RankIC: -0.060 - IR: 2.44 - Monthly win rate: 77.7% - November 2025 return: -1.52% - 12-month win rate: 58.3%[5][15] - **Smart Money Factor**: - IC: -0.037 - RankIC: -0.062 - IR: 2.72 - Monthly win rate: 81.3% - November 2025 return: 0.22% - 12-month win rate: 83.3%[5][19] - **APM Factor**: - IC: 0.028 - RankIC: 0.033 - IR: 2.23 - Monthly win rate: 76.0% - November 2025 return: -0.43% - 12-month win rate: 41.7%[5][23] - **Ideal Amplitude Factor**: - IC: -0.054 - RankIC: -0.074 - IR: 3.03 - Monthly win rate: 83.4% - November 2025 return: 0.49% - 12-month win rate: 66.7%[5][27] - **Composite Trading Behavior Factor**: - IC: 0.066 - RankIC: 0.093 - IR: 3.30 - Monthly win rate: 79.4% - November 2025 return: -0.21% - 12-month win rate: 66.7%[5][32]

- Model Name: Barra Style Factors; Model Construction Idea: The model tracks the performance of common Barra style factors; Model Construction Process: The model calculates the returns of various style factors such as market capitalization, book-to-market ratio, growth, and earnings expectations; Model Evaluation: The model provides insights into the performance of different style factors over a specific period[4][14] - Factor Name: Ideal Reversal Factor; Factor Construction Idea: The factor identifies the strongest reversal days based on the average transaction amount per trade; Factor Construction Process: 1. Retrieve the past 20 days of data for the selected stock 2. Calculate the average transaction amount per trade for each day 3. Sum the returns of the top 10 days with the highest average transaction amount (M_high) 4. Sum the returns of the bottom 10 days with the lowest average transaction amount (M_low) 5. Calculate the Ideal Reversal Factor as M = M_high - M_low 6. Repeat the above steps for all stocks to calculate their respective Ideal Reversal Factors[5][39][41] - Factor Name: Smart Money Factor; Factor Construction Idea: The factor identifies the participation of institutional investors based on minute-level price and volume data; Factor Construction Process: 1. Retrieve the past 10 days of minute-level data for the selected stock 2. Construct the indicator $St = \frac{|Rt|}{Vt^{0.25}}$, where $Rt$ is the return at minute t and $Vt$ is the volume at minute t 3. Sort the minute-level data by $St$ in descending order and select the top 20% of minutes by cumulative volume as smart money trades 4. Calculate the volume-weighted average price (VWAP) for smart money trades (VWAPsmart) 5. Calculate the VWAP for all trades (VWAPall) 6. Calculate the Smart Money Factor as $Q = \frac{VWAPsmart}{VWAPall}$[5][40][42] - Factor Name: APM Factor; Factor Construction Idea: The factor measures the difference in stock behavior between morning (or overnight) and afternoon sessions; Factor Construction Process: 1. Retrieve the past 20 days of data for the selected stock 2. Calculate the overnight and afternoon returns for the stock and the index 3. Perform a regression of the stock returns on the index returns to obtain residuals 4. Calculate the difference between overnight and afternoon residuals 5. Construct the statistic $stat = \frac{\mu(\delta_t)}{\sigma(\delta_t)/\sqrt{N}}$ 6. Perform a cross-sectional regression of the statistic on the momentum factor to obtain residuals, which are used as the APM Factor[5][41][43][44] - Factor Name: Ideal Amplitude Factor; Factor Construction Idea: The factor measures the difference in amplitude information between high and low price states; Factor Construction Process: 1. Retrieve the past 20 days of data for the selected stock 2. Calculate the daily amplitude (high price/low price - 1) 3. Calculate the average amplitude for the top 25% of days with the highest closing prices (V_high) 4. Calculate the average amplitude for the bottom 25% of days with the lowest closing prices (V_low) 5. Calculate the Ideal Amplitude Factor as V = V_high - V_low[5][46] - Composite Factor: Kaisheng Trading Behavior Composite Factor; Construction Idea: The composite factor combines multiple trading behavior factors using their ICIR values as weights; Construction Process: 1. Perform outlier removal and standardization for each trading behavior factor within the industry 2. Use the past 12 periods' ICIR values as weights to form the composite factor 3. Calculate the composite factor's returns and performance metrics[5][31] Model Backtest Results - Barra Style Factors: Market Capitalization Factor return: 1.73%, Book-to-Market Ratio Factor return: -0.31%, Growth Factor return: 0.13%, Earnings Expectations Factor return: -0.09%[4][14] Factor Backtest Results - Ideal Reversal Factor: IC: -0.050, rankIC: -0.060, IR: 2.46, Long-Short Monthly Win Rate: 77.4%, September Long-Short Return: -0.42%, Last 12 Months Long-Short Monthly Win Rate: 58.3%[6][15] - Smart Money Factor: IC: -0.037, rankIC: -0.061, IR: 2.70, Long-Short Monthly Win Rate: 81.8%, September Long-Short Return: 0.30%, Last 12 Months Long-Short Monthly Win Rate: 83.3%[6][18] - APM Factor: IC: 0.029, rankIC: 0.034, IR: 2.29, Long-Short Monthly Win Rate: 76.4%, September Long-Short Return: 1.68%, Last 12 Months Long-Short Monthly Win Rate: 50.0%[6][22] - Ideal Amplitude Factor: IC: -0.053, rankIC: -0.073, IR: 2.98, Long-Short Monthly Win Rate: 83.2%, September Long-Short Return: 0.40%, Last 12 Months Long-Short Monthly Win Rate: 66.7%[6][26] - Composite Factor: IC: 0.066, rankIC: 0.091, IR: 3.23, Long-Short Monthly Win Rate: 82.1%, September Long-Short Return: 0.57%, Last 12 Months Long-Short Monthly Win Rate: 75.0%[6][31]