Investment Rating - The investment rating for the company is "Outperform the Market" (maintained) [2][4][18] Core Views - The company positions itself as a full-stack AI service provider, significantly increasing its investment in AI infrastructure [3][5][6] - The CEO of the company highlighted three stages of AI development: intelligent emergence, autonomous action, and self-iteration, predicting a future shift towards super artificial intelligence (ASI) [3][5] - The company plans to enhance its global data center energy consumption by tenfold by 2032, indicating sustained capital expenditure (CAPEX) investments [4][6] - The company has launched several new models, including Qwen3-Max, which ranks third globally in performance, and introduced the next-generation Qwen model architecture [3][11][12] - The company aims to accelerate its international expansion by establishing data centers in Brazil, France, and the Netherlands, while also expanding operations in five other countries [3][4] Summary by Sections AI Infrastructure and Models - The company is committed to providing leading large models and cloud computing networks, with a focus on self-developed core storage systems and computing chips [4][6][13] - The company has upgraded its AI infrastructure, including high-performance servers and networks, to meet the demands of large model training and inference [13][14] Financial Projections - The revenue forecasts for FY2026, FY2027, and FY2028 are adjusted to 1,066.1 billion, 1,188.8 billion, and 1,301.4 billion respectively, reflecting a more optimistic outlook on cloud revenue growth [4][18] - Adjusted net profits for the same periods are projected at 126.7 billion, 167.2 billion, and 196.0 billion [4][18] Market Position and Strategy - The company’s full-stack self-developed capabilities and increased overseas presence are expected to improve profit margins in the long term [4][18] - The company’s strategy includes a focus on open-source models, aiming to create significant value in the LLM era [5][6]

Investment Rating - The investment rating for the company is "Outperform the Market" (maintained) [1] Core Views - The company benefits from increased upstream capital expenditure, having recently signed significant contracts in the Middle East [2][3] - The Middle East is a key area for global energy development, with Iraq and the UAE expected to maintain high upstream capital expenditures [3] - The contracts signed in Iraq and the UAE will help the company consolidate and expand its market in oil and gas transportation engineering, positively impacting revenue and profit over the next 4-5 years [3] - Revenue forecasts for 2025-2027 are projected at 899.92 billion, 944.92 billion, and 982.97 billion RMB, with net profits of 7.33 billion, 8.23 billion, and 8.50 billion RMB respectively [3][7] Summary by Sections Recent Contracts - The company’s subsidiary signed an EPC contract with Iraq's Basra Oil Company for a seawater pipeline project worth 2.524 billion USD (approximately 18.032 billion RMB) [2][5] - Another EPC contract was signed with ADNOC Gas for gas pipeline projects in the UAE, valued at 513 million USD (approximately 3.688 billion RMB) [2][7] Market Outlook - The company is well-positioned to benefit from ADNOC Gas's future investments, with ADNOC planning to invest up to 15 billion USD from 2025 to 2029 [6][7] - The GGIP project in Iraq, with an initial investment of around 10 billion USD, aims to enhance gas recovery and increase oil production, further supporting the company's growth [4] Financial Projections - The company’s earnings per share (EPS) are expected to be 0.13, 0.15, and 0.15 RMB for 2025-2027, with corresponding price-to-earnings (PE) ratios of 26.10, 22.60, and 22.60 [3][7] - The company’s financial metrics indicate a positive trend, with projected revenue growth and improved cash flow [10]

Core Insights - The report highlights the performance of various quantitative investment strategies, with a focus on the "Excellent Fund Performance Enhancement Portfolio," "Expected Selection Portfolio," "Brokerage Golden Stock Performance Enhancement Portfolio," and "Growth Stability Portfolio" [12][13][18][39]. Excellent Fund Performance Enhancement Portfolio - This portfolio aims to outperform the median return of actively managed equity funds, achieving an absolute return of 28.00% year-to-date, ranking in the 54.37th percentile among 3,469 active equity funds [23][51]. - The portfolio's performance for the week was an absolute return of 0.35%, with a relative underperformance of -0.12% compared to the mixed equity fund index [23][17]. Expected Selection Portfolio - The Expected Selection Portfolio focuses on stocks that have exceeded profit expectations, achieving a year-to-date absolute return of 46.54%, ranking in the 20.61st percentile among active equity funds [31][24]. - For the week, this portfolio recorded an absolute return of 0.70%, with a relative outperformance of 0.23% against the mixed equity fund index [31][29]. Brokerage Golden Stock Performance Enhancement Portfolio - This portfolio is constructed using stocks from the brokerage golden stock pool, achieving a year-to-date absolute return of 33.26%, ranking in the 43.07th percentile among active equity funds [38][32]. - The weekly performance showed an absolute return of -0.54%, with a relative underperformance of -1.01% compared to the mixed equity fund index [38][11]. Growth Stability Portfolio - The Growth Stability Portfolio aims to capture excess returns from growth stocks, achieving a year-to-date absolute return of 51.84%, ranking in the 15.31st percentile among active equity funds [46][39]. - For the week, this portfolio had an absolute return of 0.26%, with a relative underperformance of -0.22% against the mixed equity fund index [46][17]. Market Overview - The report indicates that the median stock return for the week was -1.74%, with 31% of stocks rising and 69% falling, while the median return for active equity funds was 0.51%, with 60% of funds rising and 40% falling [50][47]. - Year-to-date, the median stock return was 20.22%, with 81% of stocks rising and 19% falling, while the median return for active equity funds was 30.56%, with 98% of funds rising and 2% falling [50][47].

Quantitative Models and Construction Methods - **Model Name**: Guosen JinGong Hong Kong Stock Selection Portfolio **Model Construction Idea**: The model is based on a dual-layer selection process that combines fundamental and technical analysis of stocks recommended by analysts. The goal is to identify stocks with both fundamental support and technical resonance, which are expected to outperform[14][15] **Model Construction Process**: 1. Construct an analyst-recommended stock pool based on the following events: - Analysts upgrading earnings forecasts - Analysts initiating coverage - Analyst report titles indicating unexpected events 2. Perform dual-layer screening on the stock pool: - **Fundamental analysis**: Evaluate the fundamental support of the stocks - **Technical analysis**: Assess the technical resonance of the stocks 3. Select stocks that meet both fundamental and technical criteria to form the Hong Kong Stock Selection Portfolio 4. Backtesting period: 2010-01-01 to 2025-06-30, considering full position and transaction costs **Model Evaluation**: The model demonstrates strong performance with significant annualized returns and excess returns over the Hang Seng Index[15] - **Model Name**: Stable New High Stock Screening **Model Construction Idea**: The model leverages momentum and trend-following strategies, which have been shown to be effective in the Hong Kong stock market. Stocks that reach new highs are considered market leaders and are expected to deliver higher future returns[20] **Model Construction Process**: 1. Define the 250-day new high distance as: $ 250 \text{ day new high distance} = 1 - \frac{\text{Close}_{t}}{\text{ts\_max}(\text{Close}, 250)} $ - $\text{Close}_{t}$: Latest closing price - $\text{ts\_max}(\text{Close}, 250)$: Maximum closing price over the past 250 trading days - If the latest closing price reaches a new high, the 250-day new high distance is 0; if it falls from the new high, the distance is positive, indicating the degree of decline 2. Screen stocks that have reached a 250-day new high in the past 20 trading days based on the following criteria: - Analyst attention: At least 5 buy or overweight ratings in the past 6 months - Relative stock strength: Top 20% in terms of 250-day price change within the sample pool - Price stability: Evaluate using two indicators: - Price path smoothness: Stock price displacement ratio - New high persistence: Time-series average of the 250-day new high distance over the past 120 days - Trend continuation: Time-series average of the 250-day new high distance over the past 5 days 3. Select the top 50 stocks based on the above criteria, ensuring at least 50 stocks are included[22][23] Model Backtesting Results - **Guosen JinGong Hong Kong Stock Selection Portfolio**: - Annualized return: 19.11% - Excess return over Hang Seng Index: 18.48% - Maximum drawdown: 23.73% - Information ratio (IR): 1.22 - Tracking error: 14.55% - Return-to-drawdown ratio: 0.78[19] - **Stable New High Stock Screening**: - The model identified 14 stocks in the pharmaceutical sector, 11 in the cyclical sector, 8 in technology, 6 in consumer, and 4 in manufacturing as stable new high stocks for the week[22][23] Quantitative Factors and Construction Methods - **Factor Name**: 250-Day New High Distance **Factor Construction Idea**: This factor measures the distance of the latest closing price from the 250-day high, capturing the momentum and trend-following characteristics of stocks[22] **Factor Construction Process**: - Formula: $ 250 \text{ day new high distance} = 1 - \frac{\text{Close}_{t}}{\text{ts\_max}(\text{Close}, 250)} $ - $\text{Close}_{t}$: Latest closing price - $\text{ts\_max}(\text{Close}, 250)$: Maximum closing price over the past 250 trading days - Interpretation: A value of 0 indicates the stock is at its 250-day high, while a positive value indicates the percentage drop from the high[22] Factor Backtesting Results - **250-Day New High Distance Factor**: - The factor was used to identify stable new high stocks, with the pharmaceutical sector having the highest number of selected stocks (14), followed by cyclical (11), technology (8), consumer (6), and manufacturing (4)[22][23]

- The report tracks the performance of Guosen Financial Engineering's index enhancement portfolios, which are constructed based on benchmarks such as CSI 300, CSI 500, CSI 1000, and CSI A500 indices. The construction process includes three main components: return prediction, risk control, and portfolio optimization[12][14][42] - The MFE (Maximized Factor Exposure) portfolio is used to test the effectiveness of single factors under real-world constraints. The optimization model maximizes single-factor exposure while controlling for industry exposure, style exposure, stock weight deviation, and turnover rate. The objective function is defined as: $\begin{array}{ll}max&f^{T}\ w\\ s.t.&s_{l}\leq X(w-w_{b})\leq s_{h}\\ &h_{l}\leq H(w-w_{b})\leq h_{h}\\ &w_{l}\leq w-w_{b}\leq w_{h}\\ &b_{l}\leq B_{b}w\leq b_{h}\\ &\mathbf{0}\leq w\leq l\\ &\mathbf{1}^{T}\ w=1\end{array}$ Here, `f` represents factor values, `w` is the stock weight vector, and constraints include style factor exposure (`X`), industry exposure (`H`), stock weight deviation (`w`), and component stock weight control (`B_b`). The weights are normalized to sum to 1[42][43][44] - The MFE portfolio construction process involves setting constraints, optimizing the portfolio at the end of each month, and calculating historical returns during the backtesting period. Transaction costs of 0.3% are deducted on both sides to compute risk-return statistics relative to the benchmark[46] - The report monitors the performance of 30+ factors across different sample spaces, including CSI 300, CSI 500, CSI 1000, CSI A500, and public fund heavy-holding indices. Factors are categorized into valuation, reversal, growth, profitability, liquidity, governance, and analyst-related dimensions. Examples include BP (Book-to-Price), ROE (Return on Equity), and momentum factors[15][16][17] - Factor performance varies across sample spaces. For example, in the CSI 300 space, factors such as single-quarter ROE, single-quarter revenue growth, and single-quarter surprise magnitude performed well recently, while factors like BP and expected net profit growth performed poorly[18][19] - In the CSI 500 space, factors such as three-month turnover, single-quarter revenue growth, and EPTTM percentile performed well recently, while factors like one-year momentum and standardized unexpected income performed poorly[20][21] - In the CSI 1000 space, factors such as three-month institutional coverage, single-quarter ROE, and executive compensation performed well recently, while factors like one-year momentum and DELTAROA performed poorly[22][23] - In the CSI A500 space, factors such as single-quarter revenue growth, EPTTM percentile, and single-quarter ROE performed well recently, while factors like one-year momentum and DELTAROE performed poorly[24][25] - In the public fund heavy-holding index space, factors such as executive compensation, single-quarter ROE, and three-month institutional coverage performed well recently, while factors like one-year momentum and expected EPTTM performed poorly[26][27] - The report tracks the performance of public fund index enhancement products for CSI 300, CSI 500, CSI 1000, and CSI A500 indices. For example, CSI 300 index enhancement products had a maximum excess return of 0.91% and a minimum of -1.54% in the past week, with a median of -0.17%[28][32] - CSI 500 index enhancement products had a maximum excess return of 1.63% and a minimum of -1.35% in the past week, with a median of -0.01%[35] - CSI 1000 index enhancement products had a maximum excess return of 1.66% and a minimum of -0.37% in the past week, with a median of 0.44%[38] - CSI A500 index enhancement products had a maximum excess return of 0.53% and a minimum of -0.76% in the past week, with a median of -0.11%[41]

Quantitative Models and Construction Methods - **Model Name**: Maximized Factor Exposure Portfolio (MFE) **Model Construction Idea**: The MFE portfolio is designed to maximize the exposure of a single factor while controlling for various constraints such as industry exposure, style exposure, stock weight deviation, and turnover rate. This approach ensures that the factor's predictive power is tested under realistic constraints, making it more applicable in actual portfolio construction [42][43][44] **Model Construction Process**: The MFE portfolio is constructed using the following optimization model: $ \begin{array}{ll} max & f^{T} w \\ s.t. & s_{l} \leq X(w-w_{b}) \leq s_{h} \\ & h_{l} \leq H(w-w_{b}) \leq h_{h} \\ & w_{l} \leq w-w_{b} \leq w_{h} \\ & b_{l} \leq B_{b}w \leq b_{h} \\ & \mathbf{0} \leq w \leq l \\ & \mathbf{1}^{T} w = 1 \end{array} $ - **Objective Function**: Maximize single-factor exposure, where \( f \) represents factor values, \( f^{T}w \) is the weighted exposure of the portfolio to the factor, and \( w \) is the stock weight vector - **Constraints**: 1. **Style Exposure**: \( X \) is the factor exposure matrix, \( w_b \) is the benchmark weight vector, and \( s_l, s_h \) are the lower and upper bounds for style factor exposure 2. **Industry Exposure**: \( H \) is the industry exposure matrix, and \( h_l, h_h \) are the lower and upper bounds for industry deviation 3. **Stock Weight Deviation**: \( w_l, w_h \) are the lower and upper bounds for stock weight deviation 4. **Constituent Stock Weight**: \( B_b \) is a 0-1 vector indicating whether a stock is a benchmark constituent, and \( b_l, b_h \) are the lower and upper bounds for constituent stock weights 5. **No Short Selling**: Ensures non-negative weights and limits individual stock weights to \( l \) 6. **Full Investment**: Ensures the portfolio is fully invested with \( \mathbf{1}^{T}w = 1 \) - **Implementation**: 1. Set constraints for style, industry, and stock weights 2. Construct MFE portfolios for each factor at the end of each month 3. Backtest the MFE portfolios, calculate historical returns, and adjust for transaction costs (0.3% on both sides) [42][43][46] **Model Evaluation**: The MFE portfolio approach is effective in testing factor validity under realistic constraints, ensuring that factors deemed "effective" can contribute to actual portfolio performance [42][43] Quantitative Factors and Construction Methods - **Factor Name**: Single-Quarter ROE **Factor Construction Idea**: Measures the return on equity for a single quarter to capture profitability trends [17] **Factor Construction Process**: $ \text{Single-Quarter ROE} = \frac{\text{Net Income (Quarterly)} \times 2}{\text{Average Shareholders' Equity}} $ - **Net Income (Quarterly)**: Quarterly net income attributable to shareholders - **Average Shareholders' Equity**: Average of beginning and ending equity for the quarter [17] - **Factor Name**: Single-Quarter Revenue Growth (YoY) **Factor Construction Idea**: Tracks the year-over-year growth in quarterly revenue to identify growth trends [17] **Factor Construction Process**: $ \text{Single-Quarter Revenue Growth (YoY)} = \frac{\text{Revenue (Current Quarter)} - \text{Revenue (Same Quarter Last Year)}}{\text{Revenue (Same Quarter Last Year)}} $ [17] - **Factor Name**: Analyst Coverage (3-Month) **Factor Construction Idea**: Measures the number of analysts covering a stock over the past three months to gauge market attention [17] **Factor Construction Process**: $ \text{3-Month Analyst Coverage} = \text{Number of Analysts Covering the Stock in the Last 3 Months} $ [17] Factor Backtesting Results - **Single-Quarter ROE**: - **CSI 300**: Weekly excess return: 0.42%, monthly: 2.94%, YTD: 15.41%, historical annualized: 4.92% [19] - **CSI 500**: Weekly excess return: 0.47%, monthly: 0.89%, YTD: 4.43%, historical annualized: 5.85% [21] - **CSI 1000**: Weekly excess return: 1.20%, monthly: 1.70%, YTD: -0.61%, historical annualized: 7.62% [23] - **CSI A500**: Weekly excess return: 0.30%, monthly: 1.68%, YTD: 13.78%, historical annualized: 3.35% [25] - **Single-Quarter Revenue Growth (YoY)**: - **CSI 300**: Weekly excess return: 0.48%, monthly: 2.34%, YTD: 17.35%, historical annualized: 4.94% [19] - **CSI 500**: Weekly excess return: 1.28%, monthly: 2.58%, YTD: 15.18%, historical annualized: 3.81% [21] - **CSI 1000**: Weekly excess return: 0.69%, monthly: 2.73%, YTD: 15.73%, historical annualized: 5.17% [23] - **CSI A500**: Weekly excess return: 0.47%, monthly: 1.15%, YTD: 15.65%, historical annualized: 2.96% [25] - **3-Month Analyst Coverage**: - **CSI 300**: Weekly excess return: 0.17%, monthly: 0.90%, YTD: 10.33%, historical annualized: 3.07% [19] - **CSI 500**: Weekly excess return: 0.29%, monthly: 0.07%, YTD: 4.10%, historical annualized: 5.56% [21] - **CSI 1000**: Weekly excess return: 1.30%, monthly: 0.52%, YTD: 5.98%, historical annualized: 7.22% [23] - **CSI A500**: Weekly excess return: -0.21%, monthly: 0.97%, YTD: 8.12%, historical annualized: 3.93% [25]

1. Report Industry Investment Rating No relevant content provided. 2. Core Viewpoints - As market risk appetite continued to rise, the China Securities REITs Index fell 0.8% this week, with property - type REITs and franchise - type REITs averaging - 0.9% and - 0.8% respectively. All sectors closed down, with water conservancy facilities, municipal facilities, and affordable housing having the largest declines. The 9 - department joint issuance included community commercial infrastructure in the priority support scope for REITs, further expanding the underlying asset scope of REITs [1]. 3. Summary by Related Catalogs 3.1 Secondary Market Trends - **Index Performance**: As of September 26, 2025, the closing price of the China Securities REITs (closing) Index was 831.45 points, with a weekly decline of 0.8%, performing worse than the China Securities All - Bond Index (- 0.3%), the China Securities Convertible Bond Index (+ 0.9%), and the CSI 300 Index (+ 1.1%). Year - to - date, the performance order of major indices was: CSI 300 (+ 15.6%) > China Securities Convertible Bond Index (+ 15.3%) > China Securities REITs Index (+ 5.3%) > China Securities All - Bond Index (+ 0.0%). In the past year, the return of the China Securities REITs Index was 5.6% with a volatility of 7.2%. Its return was lower than that of the CSI 300 Index and the China Securities Convertible Bond Index but higher than that of the China Securities All - Bond Index; its volatility was lower than that of the CSI 300 Index and the China Securities Convertible Bond Index but higher than that of the China Securities All - Bond Index [2][7][8]. - **Market Value and Turnover**: The total market value of REITs decreased to 219 billion yuan on September 26, a decrease of 2.5 billion yuan from last week. The average daily turnover rate for the whole week was 0.33%, a decrease of 0.13 percentage points from the previous week [2][8]. - **REITs Performance by Attribute and Type**: From the perspective of different project attributes, property - type REITs and franchise - type REITs had average weekly changes of - 0.9% and - 0.8% respectively. All REITs sectors closed down, with water conservancy facilities, municipal facilities, and affordable housing having the largest declines. The top three REITs in terms of weekly gains were Bosera Jinkai Industrial Park REIT (+ 1.64%), CITIC Construction Investment SPIC New Energy REIT (+ 0.98%), and CICC Liandong Science and Technology Innovation REIT (+ 0.71%) [3][16][20]. - **Trading Activity and Fund Flow**: New infrastructure REITs had the highest trading activity this week, with an average daily turnover rate of 1.1%. Transportation infrastructure REITs had the highest trading volume ratio this week, accounting for 26.3% of the total REITs trading volume. The top three REITs in terms of net inflow of main funds were CJG First Agriculture REIT (14.67 million yuan), Southern Runze Technology Data Center REIT (10.93 million yuan), and Huaxia China Resources Commercial REIT (8.37 million yuan) [3][22][23]. 3.2 Primary Market Issuance As of September 26, 2025, there were 2 REITs products in the accepted stage, 1 in the in - inquiry stage, 3 in the feedback - received stage, 8 products that had passed and were awaiting listing, and 11 first - issued products that had passed and were listed [25]. 3.3 Valuation Tracking - **Valuation Indicators**: REITs have both bond and equity characteristics. As of September 26, the average annualized cash distribution rate of public - offering REITs was 6.2%. From the equity perspective, the relative net value premium rate, IRR, and P/FFO were used to judge the valuation of REITs. The relative net value premium rate reflects the relationship between the market value and the fair value of the fund, similar to the PB indicator of stocks; IRR is the internal rate of return calculated by the cash - flow discount method; P/FFO is the current price divided by the cash flow generated from operations [27]. - **Valuation by Project Type**: Different project types of REITs had different relative net value premium rates, P/FFO, IRR, and annualized dividend rates. For example, affordable housing had a relative net value premium rate of 44.1%, P/FFO of 34.7, IRR of 3.4%, and an annualized dividend rate of 3.3% [28]. - **Comparison with Benchmarks**: As of September 26, 2025, the dividend rate of property - type REITs was 81 basis points higher than the average dividend rate of CSI Dividend stocks, and the average internal rate of return of franchise - type REITs had a spread of 173 basis points compared with the 10 - year Treasury bond yield [30]. 3.4 Industry News - **Policy Update**: The Ministry of Commerce, together with eight other departments, issued a notice including community commercial infrastructure in the priority support scope for REITs, which helps expand the underlying asset scope of REITs and solve the capital problem in community commercial development [4][35]. - **Approved REITs Projects**: Huaxia China Overseas Commercial REIT was approved, providing a model for the transformation and upgrading of traditional real - estate enterprises. Huaxia Anbo Warehouse REIT was officially approved for issuance, with its underlying assets having obvious location advantages. Shenyang International Software Park Public - Offering REIT was approved, being the first successful public - offering REIT project in Northeast China [35].

Quantitative Models and Construction Methods - **Model Name**: Excellent Fund Performance Enhancement Portfolio **Model Construction Idea**: Transition from benchmarking broad-based indices to benchmarking active equity funds, leveraging quantitative methods to enhance fund selection and achieve "best of the best" [4][19][52] **Model Construction Process**: 1. Benchmark against the median return of active equity funds, represented by the biased equity hybrid fund index (885001.WI) [19][52] 2. Use a layered neutralization process for return-related factors to address style concentration issues [52] 3. Optimize the portfolio to control deviations in individual stocks, industries, and styles relative to the selected fund holdings [53] **Model Evaluation**: Demonstrates strong stability and the ability to consistently outperform the median of active equity funds [53] - **Model Name**: Outperformance Selection Portfolio **Model Construction Idea**: Focus on stocks with significant outperformance events, selecting those with both fundamental support and technical resonance [5][58] **Model Construction Process**: 1. Filter stocks based on research report titles indicating outperformance and analysts' upward revisions of net profit [5][58] 2. Conduct dual-layer screening on fundamentals and technicals to select stocks with both fundamental support and technical resonance [5][58] **Model Evaluation**: Consistently ranks in the top 30% of active equity funds annually, demonstrating strong performance [59] - **Model Name**: Securities Firms' Golden Stock Performance Enhancement Portfolio **Model Construction Idea**: Optimize the securities firms' golden stock pool to achieve stable outperformance relative to the biased equity hybrid fund index [6][63] **Model Construction Process**: 1. Use the securities firms' golden stock pool as the stock selection space and benchmark [6][33] 2. Optimize the portfolio to control deviations in individual stocks, industries, and styles relative to the golden stock pool [6][33] **Model Evaluation**: Consistently ranks in the top 30% of active equity funds annually, reflecting stable performance [64] - **Model Name**: Growth and Stability Portfolio **Model Construction Idea**: Focus on the "golden period" of excess returns for growth stocks, using a two-dimensional evaluation system based on time series and cross-sectional analysis [7][68] **Model Construction Process**: 1. Use the "excess return release map" to identify the strongest excess return periods before and after positive events [68] 2. Prioritize stocks closer to their financial report disclosure dates, and use multi-factor scoring to select high-quality stocks when the sample size is large [7][68] 3. Introduce mechanisms such as weak balancing, transition, buffering, and risk avoidance to reduce turnover and mitigate risks [68] **Model Evaluation**: Consistently ranks in the top 30% of active equity funds annually, with strong performance [69] --- Backtesting Results of Models - **Excellent Fund Performance Enhancement Portfolio**: - Annualized return: 20.31% - Excess return relative to biased equity hybrid fund index: 11.83% - Consistently ranks in the top 30% of active equity funds annually [54][57] - **Outperformance Selection Portfolio**: - Annualized return: 30.55% - Excess return relative to biased equity hybrid fund index: 24.68% - Consistently ranks in the top 30% of active equity funds annually [59][61] - **Securities Firms' Golden Stock Performance Enhancement Portfolio**: - Annualized return: 19.34% - Excess return relative to biased equity hybrid fund index: 14.38% - Consistently ranks in the top 30% of active equity funds annually [64][67] - **Growth and Stability Portfolio**: - Annualized return: 35.51% - Excess return relative to biased equity hybrid fund index: 26.88% - Consistently ranks in the top 30% of active equity funds annually [69][72]

Investment Rating of the Reported Industry No information provided in the given content. Core Views of the Report - In the first three quarters of 2025, the bond market yield showed an "N"-shaped trend. Credit bond yields fluctuated similarly to government bond yields, with overall wide - range volatile upward movement. Credit spreads first narrowed and then widened slightly. Default risk continued to decline, with default entities concentrated in real - estate bonds, mainly private enterprises. The amount of credit bonds with a downgraded implied rating in the ChinaBond market increased year - on - year, while the amount of upgraded ones was lower than the same period last year [9][37][38]. Summary by Relevant Catalogs Valuation Curve: Yields Fluctuated Widely and Rose - As of September 23, 2025, the yields of 1 - year Treasury bonds, 10 - year Treasury bonds, and 10 - year China Development Bank bonds changed by 30BP, 20BP, and 30BP respectively. The yields of 3 - year AAA, 3 - year AA +, 3 - year AA, and 3 - year AA - changed by 22BP, 13BP, 8BP, and - 25BP respectively. The credit spreads of 3 - year AAA, 3 - year AA +, 3 - year AA, and 3 - year AA - narrowed by 11BP, 21BP, 26BP, and 59BP respectively. Overall, the yields of medium - short - term and long - term interest - rate bonds and most credit bonds increased, and the credit spreads of various varieties narrowed, with lower - grade and shorter - term credit spreads narrowing more. The 10 - 1 curve flattened [10]. Treasury Bond Yields Presented an "N"-shaped Trend - **January - mid - March**: At the beginning of the year, the central bank suspended Treasury bond trading and reduced open - market investment to stabilize the exchange rate. The tightened capital led to an upward trend in bond market yields. After the Two Sessions, the market adjusted its expectations, and the 10 - year Treasury bond yield reached a high of 1.90% [11][12]. - **Late March - April**: The capital became looser, and the Sino - US tariff "tug - of - war" began. The 10 - year Treasury bond yield dropped to the 1.63% - 1.67% range [16]. - **May - early July**: In early May, the central bank's RRR cut and interest rate cut, along with positive results from tariff negotiations, led to a slight increase in long - end interest - rate bond yields. In June, the central bank's reverse - repurchase operations improved the capital situation, and bond yields fluctuated downward [16]. - **Mid - July - September**: The "anti - involution" policy raised inflation expectations, the equity market strengthened, and the bond market was suppressed. Bond yields rose, but short - end yields were stable, resulting in a "bear steep" pattern [16]. Credit Spreads - Credit Spreads of All Grades First Narrowed and Then Widened - **January - mid - March**: At the beginning of January, interest - rate bonds quickly adjusted upward, and credit spreads were passively narrowed. Before the Two Sessions, the market expected an RRR cut and interest rate cut, and credit spreads widened briefly. After the Two Sessions, credit spreads narrowed rapidly again [17]. - **Late March - April**: The bond market recovered quickly, and credit spreads widened slightly [17]. - **May**: Credit spreads narrowed to the lowest point of the year due to the implementation of monetary policy tools and looser capital [17]. - **June - early July**: Short - end Treasury bond yields declined, and credit spreads first widened slightly and then narrowed [17]. - **Mid - July - September**: The bond market adjusted, and credit spreads widened slightly [17]. The Risk of Downgraded Implied Rating in the ChinaBond Market Increased - In the first three quarters of 2025, the amount of credit bonds with a downgraded implied rating in the ChinaBond market was 764.1 billion, a significant year - on - year increase. The amount of upgraded credit bonds was 358 billion, significantly lower than the same period last year. The proportion of urban investment bonds in the upgraded and downgraded samples decreased both year - on - year and quarter - on - quarter [21]. Default: Default Risk Decreased, and the Default Rate of Real - Estate Bonds Declined - In the first three quarters of 2025, there were 3 new first - time default issuers. According to the broad default definition, the default amount was 6 billion, and the default rate was 0.01%, with the annualized default rate decreasing significantly compared to previous years. Default entities were mainly concentrated in real - estate bonds, mostly private enterprises. The real - estate bond default rate was 0.2%, and the default scale and annualized default rate decreased both year - on - year and quarter - on - quarter. The private enterprise default rate was 0.5%, and the annualized default rate continued to decline quarter - on - quarter [24][31]. Recovery Rate Remained Low - In the first three quarters of 2025, defaulted bonds recovered 20.76 billion in principal. From 2014 to the present, defaulted bonds have repaid 124.7 billion in principal, and the repayment rate of overdue principal was 11.9% [34].

- The report introduces a quantitative model named "250-day new high distance" to track market trends and identify hot sectors. The model is based on momentum and trend-following strategies, which have been proven effective in previous studies. The calculation formula is: $ 250\text{-day new high distance} = 1 - \frac{\text{Closet}}{\text{ts\_max(Close, 250)}} $ where Closet represents the latest closing price, and ts_max(Close, 250) is the maximum closing price over the past 250 trading days. If the latest closing price hits a new high, the distance equals 0; otherwise, it is a positive value indicating the degree of pullback[11][19][23] - The model evaluates the relative proximity of major indices to their 250-day highs. As of September 26, 2025, the distances for indices such as the Shanghai Composite Index, Shenzhen Component Index, CSI 300, CSI 500, CSI 1000, CSI 2000, ChiNext Index, and STAR 50 Index are 1.43%, 1.76%, 0.95%, 1.37%, 2.08%, 3.67%, 2.60%, and 1.60%, respectively[12][13][32] - The report also tracks the proximity of industry indices to their 250-day highs. Industries such as power equipment & new energy, non-ferrous metals, electronics, media, and machinery are closer to their 250-day highs, with distances of 1.69%, 0.82%, 2.65%, 2.74%, and 2.09%, respectively[13][15][32] - A separate analysis focuses on concept indices, including power generation equipment, gold, wind power, solar power, and energy storage, which are relatively close to their 250-day highs[15][17][32] - The report identifies 1233 stocks that reached 250-day highs in the past 20 trading days. The sectors with the highest number of such stocks are electronics (182 stocks), machinery (178 stocks), and basic chemicals (137 stocks). The sectors with the highest proportion of stocks reaching new highs are non-ferrous metals (54.84%), electronics (37.45%), and power equipment & new energy (31.09%)[19][20][33] - The report introduces a screening method for "stable new high stocks" based on factors such as analyst attention, relative stock strength, price path smoothness, and trend sustainability. Stocks are selected based on metrics like the absolute value of 120-day price changes, cumulative absolute daily price changes over 120 days, and the average 250-day new high distance over the past 120 days[23][26][33] - Using the above screening criteria, 50 stable new high stocks were identified, including companies like New Yisheng, Giant Network, and Lead Intelligent. The technology sector contributed 20 stocks, with electronics being the most represented industry. The manufacturing sector contributed 13 stocks, with machinery being the most represented industry[27][33][31]