Market Overview - The A-share market indices experienced a significant narrowing of declines this week, indicating a shift in market sentiment from panic to recovery, with a gradual rebalancing of capital styles [2][3][7] - Despite continued net outflows of main funds, a marginal improvement trend has begun to emerge, suggesting the market is in a critical window of "weak recovery + rebalancing" [2][3] Index Performance - Major A-share indices continued their adjustment but showed a notable reduction in declines compared to the previous week, transitioning from a rapid drop phase to a weak oscillation recovery phase [3][7] - Small-cap stocks experienced a technical rebound after emotional clearance, while previously resilient growth sectors, represented by the ChiNext, turned into the leading decliners, highlighting significant style rotation [3][7] Sector Rotation - The market is dominated by a "defensive + price increase" theme, with materials, utilities, and healthcare sectors rising approximately 2.5%, reflecting a preference for assets with "resource attributes + stable cash flow" [10][3] - Conversely, sectors such as information technology, finance, and certain consumer segments faced pressure, indicating that high valuation and high beta assets are still undergoing valuation digestion [10][3] Trading Volume - A-shares exhibited a "volume contraction" characteristic this week, with weekly trading volume decreasing from 11.06 trillion yuan to 10.56 trillion yuan, indicating a continued decline in trading enthusiasm [23][25] - Daily trading amounts fell from approximately 2.45 trillion yuan at the beginning of the week to 1.86 trillion yuan by Friday, with the market turnover rate dropping from 4.98% to 3.66% [23][25] Fund Flow - Main funds exhibited a "first out, then in" pattern, with a net outflow of 795 billion yuan on Monday due to external geopolitical shocks, followed by a net inflow of 150 billion yuan on Wednesday, marking a key turning point for the week [32][36] - By Friday, main funds continued to flow in with a net inflow of 82.58 billion yuan, indicating a shift from broad withdrawal to structural positioning [32][36] Market Sentiment - The market displayed a typical "V-shaped recovery" this week, with the number of stocks hitting the daily limit down reaching 145 on Monday, but quickly rebounding with a significant number of stocks hitting the limit up in subsequent days [41][46] - Margin financing balances have shown a clear downward trend, reflecting a cautious shift in sentiment, although a slight recovery was observed in the latter part of the week [42][46] Upcoming Focus - Attention will be on policy, macro data, and external disturbances, as the upcoming quarter is a crucial window for assessing economic recovery [50][51] - The market will also face the unlocking of restricted shares for 26 companies next week, which may exert pressure on stock prices [51][53]

Core Viewpoint - The article discusses the performance of dividend stocks compared to the CSI 300 index over the past 20 years, highlighting that while dividend stocks have shown resilience in recent years, their long-term performance during market downturns has been similar to that of the CSI 300 index [3][19]. Group 1: Historical Performance Comparison - Over a 20-year period, both the CSI Dividend Index and the CSI 300 experienced similar maximum drawdowns, with the dividend index at -72.13% and the CSI 300 at -72.30%, indicating almost identical performance during market declines [7][8]. - The correlation of monthly returns between the two indices was found to be 0.932, suggesting a high degree of synchronization in their performance [7]. Group 2: Recent Trends and Changes - In the last five years, the performance of dividend stocks has diverged from the CSI 300, with the maximum drawdown for dividend stocks being only -21%, while the CSI 300 faced a -46% decline [12]. - The correlation between the two indices dropped significantly to 0.44 in 2021, indicating a structural change in market dynamics, where funds shifted towards dividend stocks during periods of growth stock declines [12][14]. Group 3: Investment Strategy Insights - The article emphasizes that the recent outperformance of dividend stocks is not due to a fundamental change in their nature but rather a result of increased style differentiation between growth and dividend stocks [14]. - Historical data suggests that while dividend stocks have provided higher returns through reinvested dividends, they still require active management to optimize performance, as evidenced by the difference in annualized returns between passive holding and trend-based strategies [17][18].

Quantitative Models and Construction Methods 1. Model Name: "百元转股溢价率" (Premium per 100 Yuan Conversion Value) - **Model Construction Idea**: This model compares the valuation of convertible bonds and their underlying stocks by constructing a time-series comparable valuation metric, "百元转股溢价率" (Premium per 100 Yuan Conversion Value) [3][12] - **Model Construction Process**: 1. Fit a cross-sectional relationship curve between the conversion premium and conversion value for each time point 2. Substitute a conversion value of 100 into the fitted formula to calculate the "百元转股溢价率" 3. Formula: $$ y_{i} = \alpha_{0} + \alpha_{1} \cdot \frac{1}{x_{i}} + \epsilon_{i} $$ - \( y_{i} \): Conversion premium of the \( i \)-th bond - \( x_{i} \): Conversion value of the \( i \)-th bond [40] - **Model Evaluation**: Provides a relative valuation metric for comparing convertible bonds and stocks [3][12] 2. Model Name: "修正 YTM – 信用债 YTM" (Adjusted YTM – Credit Bond YTM) - **Model Construction Idea**: This model isolates the impact of conversion terms on the yield-to-maturity (YTM) of convertible bonds to assess the relative valuation between debt-heavy convertible bonds and credit bonds [4][12] - **Model Construction Process**: 1. Adjust the YTM of debt-heavy convertible bonds by accounting for the probability of conversion 2. Formula: $$ \text{Adjusted YTM} = \text{Convertible Bond YTM} \times (1 - \text{Conversion Probability}) + \text{Expected Annualized Return from Conversion} \times \text{Conversion Probability} $$ 3. Use the Black-Scholes (BS) model to calculate the conversion probability \( N(d2) \) 4. Calculate the median difference between adjusted YTM and credit bond YTM: $$ \text{"Adjusted YTM – Credit Bond YTM Median"} = \text{median}\{X_1, X_2, ..., X_n\} $$ - \( X_i \): Difference between adjusted YTM and credit bond YTM for the \( i \)-th bond [41] - **Model Evaluation**: Highlights the lower cost-effectiveness of debt-heavy convertible bonds compared to credit bonds [4][13] --- Model Backtesting Results 1. "百元转股溢价率" - Rolling 3-year percentile: 93.50% [3][13] - Rolling 5-year percentile: 96.10% [3][13] 2. "修正 YTM – 信用债 YTM" - Median value: -5.12% [4][13] --- Quantitative Factors and Construction Methods 1. Factor Name: 转股溢价率偏离度 (Conversion Premium Deviation) - **Factor Construction Idea**: Measures the deviation of the conversion premium from its fitted value to assess valuation differences across bonds with varying parities [18] - **Factor Construction Process**: 1. Formula: $$ \text{Conversion Premium Deviation} = \text{Conversion Premium} - \text{Fitted Conversion Premium} $$ 2. The fitted conversion premium is determined by the number of bonds, which affects the fitting quality [18] - **Factor Evaluation**: Effective in identifying valuation anomalies across bonds [18] 2. Factor Name: 理论价值偏离度 (Theoretical Value Deviation) - **Factor Construction Idea**: Measures the price expectation gap by comparing the closing price of a bond to its theoretical value, calculated using Monte Carlo simulation [18] - **Factor Construction Process**: 1. Formula: $$ \text{Theoretical Value Deviation} = \frac{\text{Convertible Bond Closing Price}}{\text{Theoretical Value}} - 1 $$ 2. Monte Carlo simulation considers conversion, redemption, downward revision, and resale clauses, simulating 10,000 paths at each time point using the same credit term limit rate as the discount rate [18] - **Factor Evaluation**: Particularly effective for equity-heavy convertible bonds [17][18] 3. Composite Factor Name: 转债综合估值因子 (Convertible Bond Comprehensive Valuation Factor) - **Factor Construction Idea**: Combines the rankings of the two factors above to create a comprehensive valuation metric [18] - **Factor Construction Process**: 1. Formula: $$ \text{Convertible Bond Comprehensive Valuation Factor} = \text{Rank}(\text{Conversion Premium Deviation}) + \text{Rank}(\text{Theoretical Value Deviation}) $$ 2. Bonds are ranked based on their factor scores, and the top 1/3 are selected to construct low-valuation indices for equity-heavy, balanced, and debt-heavy convertible bonds [19] - **Factor Evaluation**: Demonstrates strong performance across different bond categories [17][18] --- Factor Backtesting Results 1. Low-Valuation Factors (Equity-Heavy, Balanced, Debt-Heavy) - **1-Month Excess Returns**: - Equity-heavy: 4.73% - Balanced: 2.70% - Debt-heavy: -0.05% [5][20] - **Annualized Metrics (2018-02-14 to 2026-03-13)**: | Index Type | Annualized Return | Annualized Volatility | Max Drawdown | IR | Calmar Ratio | |--------------------|--------------------|------------------------|--------------|-------|--------------| | Equity-Heavy Index | 26.32% | 20.74% | 0.23 | 1.27 | 1.15 | | Balanced Index | 15.87% | 12.01% | 0.16 | 1.32 | 0.99 | | Debt-Heavy Index | 12.38% | 9.77% | 0.18 | 1.27 | 0.70 | [21] 2. Convertible Bond Style Rotation - **1-Month Return**: -0.65% - **YTD Return (2026)**: 9.34% - **Annualized Metrics (2018-02-14 to 2026-03-13)**: | Strategy Type | Annualized Return | Annualized Volatility | Max Drawdown | IR | Calmar Ratio | |----------------------------|--------------------|------------------------|--------------|-------|--------------| | Style Rotation Strategy | 25.60% | 16.95% | 15.89% | 1.51 | 1.61 | | Low-Valuation Equal Weight | 14.95% | 11.37% | 15.48% | 1.32 | 0.97 | | Equal Weight Index | 10.55% | 12.08% | 20.60% | 0.87 | 0.51 | [26]

Core Viewpoint - The article introduces the newly published book "Personal Pension Investment Guide," highlighting its relevance in the context of an aging society and the importance of personal pension planning. It emphasizes the benefits of personal pension accounts, including tax deferral advantages and investment options available within these accounts [6][58]. Group 1: Book Launch and Benefits - The book "Personal Pension Investment Guide" has been launched and quickly topped sales charts on platforms like JD.com [3]. - The book is designed for individuals looking to invest in personal pension accounts and includes a limited-time welfare package with three benefits for early purchasers [5]. - The book aims to assist investors in understanding personal pensions and making informed investment decisions [9]. Group 2: Personal Pension Account Overview - Personal pension accounts are part of the third pillar of retirement savings, allowing individuals to save independently for retirement beyond state and corporate pensions [11]. - Individuals can contribute up to 12,000 yuan annually to their personal pension accounts, which can be deducted from taxable income, with a lower tax rate of 3% applicable upon withdrawal during retirement [11][12]. - The tax deferral benefits are particularly advantageous for individuals with higher income and tax rates [14][19]. Group 3: Target Audience for Personal Pension Accounts - The personal pension accounts are particularly suitable for individuals with low pension replacement rates, those seeking a better quality of life in retirement, parents of only children, and those who wish to retire early [8][19]. Group 4: Investment Options - The personal pension accounts allow investments in five categories: commercial pension insurance, savings deposits, wealth management products, public funds, and government bonds [20]. - As of December 31, 2025, there are 309 FOF funds and 96 index funds included in the personal pension fund directory [21]. Group 5: Investment Strategies - The recommended investment strategy for ordinary investors involves a combination of leading strategy (A-series indices) and dividend strategy funds [25][58]. - The article discusses the importance of diversifying investments across different styles (growth and value) and the need for rebalancing to achieve stable excess returns [33]. Group 6: Tax Benefits and Management - Contributions to personal pension accounts qualify for tax deductions, and individuals can manage their tax liabilities through the personal income tax app [49][50]. - The article explains the process of annual tax reconciliation, allowing individuals to adjust their tax payments based on contributions to personal pension accounts [51][52].

Core Viewpoint - The article discusses strategies for identifying undervalued investment opportunities, managing growth/value style rotations, and the benefits of diversified allocation and rebalancing in investment portfolios [1][5]. Group 1: Low Valuation Investment - Investing during undervaluation is a key principle in stock fund investment, as it can reduce volatility risk and provide greater future valuation upside [7][8]. - The "Screw Nut Star Rating" is introduced as a tool to assess overall market valuation, updated daily on the public account [8][9]. - The star rating system ranges from 1 to 5 stars, indicating different investment phases, with 5 stars representing the best investment conditions [13]. Group 2: Stock Fund Portfolio Construction - The article emphasizes the importance of constructing a stable stock fund portfolio, especially during market volatility, such as conflicts in the Middle East [5]. - It suggests that stocks are suitable for investment during 4-5 star ratings, as historical data shows higher returns and lower risks when investing in undervalued phases [15][16]. Group 3: Diversified Allocation and Rebalancing - Diversified allocation is crucial due to the style rotation characteristic of the A-share market, where different styles do not move in tandem [25][26]. - Historical performance shows that diversifying between growth and value styles can lower overall portfolio volatility [26]. - A classic pairing strategy is proposed, combining leading strategies (A-series indices) with dividend strategies, which has shown to yield higher returns with lower volatility compared to the CSI 300 [32]. Group 4: Active Management and Dynamic Adjustment - The article discusses the active management of investment styles, where adjustments are made based on market conditions and valuation levels [39][42]. - For instance, in early 2026, a shift was made to reduce exposure to overvalued growth stocks while increasing investment in undervalued value stocks [43]. - The article also notes that the strategy of diversified allocation and rebalancing may face challenges during periods of high overall market valuation or liquidity constraints [47][50]. Group 5: Practical Application - The "Screw Nut Personal Pension Investment Plan" is highlighted as a practical application of the discussed strategies, focusing on a mix of leading and dividend strategies [34][51]. - The article encourages readers to utilize the daily updates on the "Screw Nut Star Rating" and index valuation tables to identify undervalued investment opportunities [52].

Quantitative Models and Construction Methods - **Model Name**: Short-term Timing Strategy **Construction Idea**: The model integrates macroeconomic, valuation, sentiment, and liquidity signals to generate short-term timing recommendations[12][13][15] **Construction Process**: 1. **Macroeconomic Signal**: - Manufacturing PMI < 50 indicates weak economic activity, giving cautious signals - Credit pulse growth at 91.53% percentile over the past 5 years indicates strong credit growth, giving optimistic signals - M1 growth rate at 91.53% percentile over the past 5 years indicates strong monetary expansion, giving optimistic signals[12][15] 2. **Valuation Signal**: - PE median at 98.35% percentile over the past 5 years suggests high valuation, giving cautious signals - PB median at 97.27% percentile over the past 5 years suggests high valuation, giving cautious signals[12][15] 3. **Sentiment Signal**: - Beta dispersion at 16.95% percentile indicates low market risk, giving optimistic signals - Volume sentiment score at 47.73% percentile indicates neutral sentiment - Volatility at 66.34% percentile indicates neutral sentiment[13][15] 4. **Liquidity Signal**: - Money market rate at 33.90% percentile indicates loose liquidity, giving optimistic signals - Exchange rate expectation at 8.47% percentile indicates strong RMB, giving optimistic signals - 5-day average net financing at 19.69% percentile indicates neutral leverage signals[13][15] **Evaluation**: The model demonstrates consistent performance with reduced drawdowns compared to benchmarks, indicating robustness[14][18] - **Model Name**: Growth-Value Style Rotation Model **Construction Idea**: The model evaluates macroeconomic, valuation, momentum, breadth, and crowding signals to recommend growth or value style allocations[22][23] **Construction Process**: 1. **Dynamic Macro Signal**: Currently neutral due to weak credit pulse and structural headwinds for growth stocks[22][23] 2. **Valuation Signal**: Growth stocks supported by valuation but weaker than value stocks in terms of price-volume trends and PB dispersion[22][23] 3. **Momentum Signal**: Short-term momentum signals favor value stocks over growth stocks[23] 4. **Breadth Signal**: Breadth indicators show stronger performance for value stocks[23] 5. **Crowding Signal**: Crowding metrics favor value stocks due to lower risk of over-concentration[23] **Evaluation**: The model has delivered annualized excess returns of 7.97% since 2011, outperforming benchmarks significantly[22][23] - **Model Name**: Small-Cap vs Large-Cap Style Rotation Model **Construction Idea**: The model uses 11 effective rotation indicators to construct a composite signal for small-cap and large-cap allocations[25][27] **Construction Process**: 1. **Indicators**: - A-share leaderboard buying intensity - R007 interbank rate - Financing balance changes - Thematic investment sentiment - Credit spread - Option volatility risk premium - Beta dispersion - PB divergence - Block trade premium/discount rate - MACD for CSI 1000 - CSI 1000 trading volume[27] 2. **Signal Aggregation**: Composite signals are smoothed over 3 days to reduce noise and provide stable recommendations[27] **Evaluation**: The model has consistently generated positive excess returns annually since 2014, demonstrating strong predictive power[26][27] Model Backtesting Results - **Short-term Timing Strategy**: - Annualized return: 16.32% - Annualized excess return: 11.34% - Maximum drawdown: 15.05% - Sharpe ratio: 0.9617 - Monthly win rate: 65.63% - Annual win rate: 80.00%[14][18][19] - **Growth-Value Style Rotation Model**: - Annualized return: 14.63% - Annualized excess return: 7.97% - Maximum drawdown: 40.08% - Sharpe ratio: 0.65 - Monthly excess win rate: 66.32% - Annualized tracking error: 5.88% - Information ratio (IR): 1.35[22][23] - **Small-Cap vs Large-Cap Style Rotation Model**: - Annualized return: 20.23% - Annualized excess return: 12.73% - Maximum drawdown: 40.70% - Monthly win rate: 50.00% - Annualized tracking error: 9.58% - Information ratio (IR): 1.33[26][27]

Quantitative Models and Construction Methods 1. Model Name: Broad-based Index Timing Strategy - **Model Construction Idea**: Utilize macroeconomic variable combinations to predict the future returns of the CSI All Share Index. The strategy involves making long or short decisions based on the predicted values exceeding a threshold[31][34]. - **Model Construction Process**: 1. Combine liquidity and inventory sub-strategies to predict whether the CSI All Share Index will rise. 2. If any predicted value exceeds the threshold (0.6), go long on the CSI All Share Index; otherwise, go short[31]. - **Model Evaluation**: The model effectively captures the impact of macroeconomic variables on the index, providing a systematic approach to timing[34]. 2. Model Name: Dividend Index Timing Strategy - **Model Construction Idea**: Use combinations of inflation and inventory, as well as inventory and credit, to predict the future returns of the Dividend Index. The strategy involves making long or short decisions based on the average predicted values exceeding a threshold[40]. - **Model Construction Process**: 1. Calculate the average predicted value of inflation + inventory and inventory + credit sub-strategies. 2. If the average exceeds the threshold (0.6), go long on the Dividend Index; otherwise, go short[40]. - **Model Evaluation**: The model demonstrates strong defensive characteristics of the Dividend Index, particularly under specific macroeconomic conditions[40]. 3. Model Name: Style Rotation Strategy - **Model Construction Idea**: Leverage macroeconomic factor combinations to predict the future returns of six style indices. Allocate capital to the top two indices with the highest predicted returns[49][54]. - **Model Construction Process**: 1. Use combinations of inflation + inventory and inflation + credit to predict the future returns of six style indices. 2. Smooth the predicted returns and rank them at the end of each month. 3. Allocate capital equally to the top two indices for the next month[54]. - **Model Evaluation**: The strategy effectively captures the differentiated impacts of macroeconomic factors on various styles, providing a robust framework for style rotation[49][54]. --- Model Backtesting Results 1. Broad-based Index Timing Strategy - **Annualized Return**: 15.34% - **Annualized Volatility**: 22.02% - **Sharpe Ratio**: 0.74 - **Maximum Drawdown**: -28.10% - **Excess Return**: 10.31% - **Tracking Error**: 34.16% - **IR**: 0.30 - **Relative Maximum Drawdown**: -50.30%[36]. 2. Dividend Index Timing Strategy - **Annualized Return**: 10.32% - **Annualized Volatility**: 13.74% - **Sharpe Ratio**: 0.75 - **Maximum Drawdown**: -19.92% - **Excess Return**: 7.97% - **Tracking Error**: 9.23% - **IR**: 0.86 - **Relative Maximum Drawdown**: -12.47%[42]. 3. Style Rotation Strategy - **Annualized Return**: 14.79% - **Annualized Volatility**: 21.81% - **Sharpe Ratio**: 0.64 - **Maximum Drawdown**: -45.93% - **Excess Return**: 4.61% - **Tracking Error**: 10.28% - **IR**: 0.52 - **Relative Maximum Drawdown**: -81.71%[59]. --- Quantitative Factors and Construction Methods 1. Factor Name: Macroeconomic Factor Variables - **Factor Construction Idea**: Select significant macroeconomic sub-variables through regression analysis and weight them inversely by their standard deviation over the past year. Use HP filter to adjust for short-term fluctuations and identify long-term trends[2]. - **Factor Construction Process**: 1. Perform regression of macroeconomic indices against broad-based indices and proxy macroeconomic variables. 2. Select sub-variables with significant t-values. 3. Weight the selected variables inversely by their past-year standard deviation. 4. Apply a one-sided HP filter to remove short-term noise and identify long-term trends[2]. - **Factor Evaluation**: The factor construction process effectively integrates macroeconomic trends and states, providing a comprehensive framework for understanding asset price drivers[2]. 2. Factor Name: High-dimensional Macroeconomic Variables - **Factor Construction Idea**: Combine marginal changes and states of macroeconomic variables to address inconsistencies in traditional macroeconomic factor transmission[2][8]. - **Factor Construction Process**: 1. Identify five dimensions of macroeconomic variables: economic prosperity, inflation, interest rates, inventory, and credit. 2. Combine marginal changes and time-series rankings of these variables to construct high-dimensional macroeconomic factors[9]. - **Factor Evaluation**: The high-dimensional approach improves the stability and predictive power of macroeconomic factors, addressing the limitations of single-dimensional indicators[8][9]. --- Factor Backtesting Results 1. Macroeconomic Factor Variables - **Liquidity (Up)**: 70.30% probability of index rise - **Liquidity (Down)**: 58.33% probability of index rise - **Inventory (Up)**: 65.84% probability of index rise - **Inventory (Down)**: 63.91% probability of index rise[37]. 2. High-dimensional Macroeconomic Variables - **Inflation (Up)**: 58.91% probability of index rise - **Inflation (Down)**: 67.33% probability of index rise - **Inventory (Up)**: 64.13% probability of index rise - **Inventory (Down)**: 63.91% probability of index rise[47].

1. Report Industry Investment Rating The provided content does not include information about the industry investment rating, so this part is skipped. 2. Core Viewpoints of the Report - The report constructs multiple ETF portfolios, including macro industry portfolio, macro + momentum industry portfolio, core - satellite portfolio, and trinity style rotation ETF portfolio, aiming to find potential investment opportunities and manage risks [1][5]. - The macro industry portfolio selects ETFs based on the sensitivity scores of economy, liquidity, and credit, and currently leans towards TMT and innovative drugs [1][7]. - The macro + momentum industry portfolio combines macro and momentum methods, with a relatively high proportion of cyclical industries selected by the momentum approach [1][14]. - The core - satellite portfolio uses the CSI 300 as the core and combines it with industry and Smart Beta portfolios, showing relatively stable performance [21]. - The trinity style rotation ETF portfolio constructs a style rotation model centered on macro - liquidity, and the current model leans towards the small - cap growth - high - quality part [6][29]. 3. Summary According to the Directory 3.1 ETF Portfolio Construction Methods 3.1.1 ETF Portfolio Construction Based on Macro - Methods - Calculate the macro - sensitivity of the indices tracked by broad - based, industry - themed, and Smart Beta ETFs according to economic, liquidity, and credit variables, and select ETFs monthly based on the current macro - variable status and index macro - sensitivity [5]. - Traditional cyclical industries are sensitive to the economy, TMT is sensitive to liquidity and insensitive to the economy, and consumption is relatively sensitive to credit. State - owned enterprises and ESG - related themes have low sensitivity to liquidity and credit [5]. - Three ETF portfolios, namely the macro industry portfolio, macro + momentum industry portfolio, and core - satellite industry portfolio, are constructed and rebalanced monthly [5]. 3.1.2 Trinity Style Rotation ETF Portfolio Construction - Build a medium - to long - term style rotation model centered on macro - liquidity, and compare it with the CSI 300 index [6]. - Construct three types of models: growth/value rotation model, market - cap model, and quality model. Combine the results of the three models to get the final style preference, with a total of 8 style preference results [6]. - Select ETFs with high exposure to the target style, control the industry exposure of ETFs to be similar to the style portfolio, and set the allocation upper and lower limits of 3% - 20% to obtain the ETF allocation model [6]. 3.2 Macro Industry Portfolio - Select industry - themed indices tracked by ETFs that have been established for more than 1 year and have a current scale of over 200 million. Calculate the sensitivity scores of economy, liquidity, and credit monthly, adjust the score directions according to the latest economic, liquidity, and credit judgment indicators, and sum them up. If liquidity and credit deviate significantly, remove the liquidity score. Select the top 6 industry - themed indices and allocate the corresponding largest - scale ETFs equally [7][8]. - Currently, the economy's leading indicators are falling, liquidity is loose, and credit indicators are tightened. The portfolio is configured with ETFs that are insensitive to the economy, sensitive to liquidity, and insensitive to credit, mainly focusing on TMT and innovative drugs. The March positions include ETFs such as GF China Hong Kong Innovative Drugs ETF and Huaxia CSI 5G Communication Theme ETF [12]. - The portfolio has relatively large fluctuations, and the excess return declined in February [13]. 3.3 Macro + Momentum Industry Portfolio - Combine the macro and momentum methods to form a complementary relationship. The momentum method first groups industry themes into 6 different groups using clustering, and then selects the product with the highest increase in the past 6 months from each group for equal - weight allocation [14]. - The industries selected by the momentum method still have a relatively high proportion of cyclical industries. The March positions include ETFs such as GF China Hong Kong Innovative Drugs ETF and Cathay CSI Semiconductor Materials and Equipment Theme ETF [18]. - The portfolio has performed well this year and continued to outperform in February [19]. 3.4 Core - Satellite Portfolio - Due to the high volatility of industry - themed ETFs and the accelerated industry rotation in the past two years, a "core - satellite" portfolio with the CSI 300 as the core is designed [21]. - Use the macro - sensitivity measurement method to measure the three ETF - tracking index pools of domestic broad - based, industry - themed, and Smart Beta ETFs, construct three stock portfolios, and then weight them at 50%, 30%, and 20% to obtain the final core + satellite portfolio [21]. - The current configuration of broad - based ETFs leans towards the science and technology innovation board and the ChiNext board. The portfolio has performed stably, outperforming in most months except for December, and continued to outperform in February 2026 [26][28]. 3.5 Trinity Style Rotation ETF Portfolio - The current model leans towards the small - cap growth - high - quality part. The factor exposure and historical performance of the model are provided, including factors such as the bond futures - spot spread, US one - year Treasury yield, and trading volume of the Shanghai and Shenzhen stock markets [29][30]. - The March positions include ETFs such as Invesco Great Wall CSI Hong Kong Stock Connect Technology ETF and Invesco Great Wall CSI Guoxin Hong Kong Stock Connect Central State - owned Enterprise Dividend ETF [35]. - The portfolio has achieved certain excess returns in many months [33].

Market Overview - In February 2026, the domestic stock market indices showed mixed performance, with the CSI 1000 index rising significantly by 3.71%, while the STAR 50 index fell by 1.42%. Year-to-date, the CSI 500 index performed strongly with a gain of 15.98%, whereas the CSI 300 index lagged with a rise of only 1.74% [2][6]. - In terms of market style, small-cap stocks outperformed large-cap stocks in February, and value stocks slightly outperformed growth stocks. Year-to-date, the ChiNext index rose by 3.34%, while the CSI 1000 index increased by 12.71% [2][7]. - The best-performing sectors in February were steel (+9.52%), building materials (+7.72%), and machinery (+7.56%), while the worst performers were media (-4.22%), non-bank financials (-3.48%), and consumer services (-3.37%) [2][12]. - The bond market saw the China Bond Treasury Total Wealth Index increase by 0.26% and the China Bond Corporate Bond Total Wealth Index rise by 0.23% in February. Year-to-date, government bonds performed slightly better than credit bonds [2][7]. - In the commodity market, most domestic commodity indices fell in February, with the Nanhua Agricultural Products Index rising by 0.23% and the Nanhua Gold Index declining by 1.17%. Year-to-date, the Nanhua Gold Index increased by 17.11% [2][7]. ETF Strategy Performance - The report highlights the growing importance of ETFs as investment tools, with a focus on various strategies including style rotation, quantitative selection, global allocation, bond allocation, and asset allocation. The current ETF combinations include style rotation, global allocation, valuation-selected ETFs, and others [3][13]. - As of February 28, 2026, the Style Rotation Portfolio achieved a cumulative return of 104.86%, outperforming its benchmark by 53.29%. The annualized volatility was 19.24%, with a maximum drawdown of -22.20% [3][19]. - The 28 Rotation Portfolio recorded a cumulative return of 66.89%, exceeding its benchmark by 17.31%, with an annualized volatility of 11.26% and a maximum drawdown of -16.39% [3][26]. - The Valuation-Selected ETF Portfolio achieved a cumulative return of 190.69%, outperforming its benchmark by 141.54%, with an annualized volatility of 20.37% and a maximum drawdown of -21.42% [3][33]. - The Global Allocation Portfolio had a cumulative return of 73.57%, surpassing its benchmark by 35.07%, with an annualized volatility of 13.56% and a maximum drawdown of -28.69% [3][41]. - The Dynamic Duration Strategy Portfolio achieved a cumulative return of 20.44%, outperforming its benchmark by 3.32%, with an annualized volatility of 1.71% and a maximum drawdown of -2.38% [3][49]. - The Asset Rotation Strategy Portfolio recorded a cumulative return of 78.61%, exceeding its benchmark by 54.43%, with an annualized volatility of 10.73% and a maximum drawdown of -12.17% [3][56]. - The Asset Rotation Strategy 2.0 Portfolio achieved a cumulative return of 75.02%, outperforming its benchmark by 50.83%, with an annualized volatility of 7.44% and a maximum drawdown of -7.95% [3][64].

- The short-term timing model maintained a neutral signal this week, with macro fundamentals neutral, valuation cautious, sentiment neutral, and liquidity optimistic[14][15][16] - The short-term timing strategy achieved an annualized return of 16.37% since 2012, with a benchmark annualized return of 5.01%, generating an annualized excess return of 11.36%. The strategy's maximum drawdown was only 15.05%, significantly better than the benchmark strategy[17][19] - Since 2024, the short-term timing strategy achieved an annualized return of 28.07%, with a benchmark annualized return of 10.18%, generating an annualized excess return of 17.89%. The strategy's maximum drawdown was 11.04%, with a Sharpe ratio of 1.4643[20] - The growth-value rotation model currently recommends overweighting value stocks. Although mid-term trends slightly support growth stocks, unfavorable macro fundamentals and weaker short-term price-volume trends suggest reducing risk by overweighting value stocks[22] - The growth-value rotation strategy achieved an annualized return of 14.62% since 2011, with a benchmark annualized return of 6.65%, generating an annualized excess return of 7.98%. This year, the strategy's excess return was 2.08%[22][23] - The small-cap and large-cap rotation model recommends overweighting large-cap stocks due to weakening small-cap price-volume signals. The strategy has generated positive annual excess returns since 2014[25][26] - The small-cap and large-cap rotation strategy achieved an annualized return of 20.35%, with an annualized excess return of 12.83%, a maximum drawdown of 40.70%, and a monthly win rate of 50.21%[27]