金融工程专题研究：华夏中证2000ETF投资价值分析：布局小微盘行情的超额之选

Quantitative Models and Factor Analysis Quantitative Models and Construction - Model Name: Brinson Attribution Model Model Construction Idea: The model is used to decompose the excess return of a portfolio relative to its benchmark into contributions from sector allocation and stock selection [59] Model Construction Process: The Brinson model decomposes the excess return ($R_p - R_b$) into two components: $R_p - R_b = \sum_{i=1}^{N} (w_{pi} - w_{bi})R_{bi} + \sum_{i=1}^{N} w_{pi}(R_{pi} - R_{bi})$ - $w_{pi}$: Portfolio weight in sector $i$ - $w_{bi}$: Benchmark weight in sector $i$ - $R_{pi}$: Portfolio return in sector $i$ - $R_{bi}$: Benchmark return in sector $i$ The first term represents the sector allocation effect, and the second term represents the stock selection effect [59] Model Evaluation: The model effectively identifies the primary source of excess returns, highlighting the significant contribution of stock selection in the analyzed ETF [59] Quantitative Factors and Construction - Factor Name: Sampling Replication Factor Factor Construction Idea: This factor aims to balance tracking error and liquidity by selecting representative stocks from the index components [51][54] Factor Construction Process: - Layered Sampling Method: - Divide stocks into groups based on characteristics such as market capitalization and valuation - Select representative stocks from each group to form the portfolio [51] - Optimization Sampling Method: - Minimize tracking error by controlling the portfolio's risk exposure to match the benchmark index - Use optimization techniques to ensure the portfolio aligns with the index's style factors (e.g., size, valuation, momentum) [54] Factor Evaluation: The factor ensures efficient tracking of the benchmark index while maintaining liquidity, making it suitable for indices with a large number of components [51][54] Backtesting Results of Models and Factors - Brinson Attribution Model: - Excess Return Attribution (2023-12-31): Total Excess Return: 3.09%, Sector Allocation: -0.04%, Stock Selection: 3.13% [60] - Excess Return Attribution (2024-06-30): Total Excess Return: 0.50%, Sector Allocation: -0.15%, Stock Selection: 0.65% [60] - Excess Return Attribution (2024-12-31): Total Excess Return: 3.10%, Sector Allocation: 0.10%, Stock Selection: 3.00% [60] - Sampling Replication Factor: - Tracking Error Metrics: - Average Deviation from Benchmark: 0.16%-0.44% - Maximum Deviation from Benchmark: 0.43%-1.35% [60] - Stock Coverage: - Number of Holdings (2023 Year-End): 824 stocks (812 from the index) - Number of Holdings (2024 Mid-Year): 844 stocks (837 from the index) - Number of Holdings (2024 Year-End): 914 stocks (901 from the index) [54] - ETF Performance Metrics: - Annualized Excess Return: 15.14% - Annualized Information Ratio (IR): 3.14 [56][58] - Cumulative Excess Return Since Inception: 30.46% [56] - IPO Allocation Returns: 2024: 1.80%, 2025 H1: 0.99% [62]