人形机器人商业化进程持续加速，消息面上，机器人龙头企业优必选于昨日宣布再签3000万元人形机器人大单，企业总订单金额已逼近4.3亿元；开普勒机 器人于近日正式发布K2大黄蜂量产启动视频，宣告全球首款商业可售的混动架构人形机器人开始交付商业订单。 机器人板块今日延续涨势，截至收盘，中证智能电动汽车指数上涨1.8%，中证消费电子主题指数上涨1.6%，国证机器人产业指数上涨1.1%，中证物联网主 题指数上涨0.9%。机器人ETF易方达（159530）全天获超6000万份净申购。 | 令日 该指数涨跌 | 该指数 该指数自2020年 滚动市盈率 发布以来估值分们 | | --- | --- | | 1.6% | 66.7倍 99.8% | （文章来源：每日经济新闻） ...

证券研究报告：金融工程报告 发布时间：2025-09-28 研究所 分析师:黄子崟 SAC 登记编号:S1340523090002 Email:huangziyin@cnpsec.com 研究助理:李子凯 SAC 登记编号:S1340124100014 Email:lizikai@cnpsec.com 近期研究报告 《指数震荡反内卷方向领涨，ETF 持续 净流入金融地产——行业轮动周报 20250921》 - 2025.09.22 《成长风格占优，小盘股活跃——中邮 因子周报 20250914》 - 2025.09.15 《金融地产获 ETF 持续净流入，连板情 绪偏修复等待合力方向——行业轮动 周报 20250914》 - 2025.09.15 《融资资金净流入电力设备及新能源， 光通信概念趋势未破——行业轮动周 报 20250907》 - 2025.09.08 《双创涨速明显提升，ETF 资金配置思 路偏补涨——行业轮动周报 20250831》 - 2025.09.01 《指数上行重返十年高位，涨幅超 10% 芯片相关 ETF 净流出较多——行业轮动 周报 20250824》 – 2025.08.25 《 ...

Core Viewpoint - The stock ETF market experienced a net inflow of approximately 8 billion yuan on September 23, with significant inflows into semiconductor, securities, artificial intelligence, and robotics sector ETFs, while broad-based ETFs like the CSI 300 Index and ChiNext Index saw substantial outflows [2][3][6]. Fund Flow Summary - On September 23, the total scale of all stock ETFs reached 4.4 trillion yuan, with 1,213 stock ETFs (including cross-border ETFs) [3]. - The top three ETFs by net inflow were the Jiashi Sci-Tech Chip ETF, Guotai Securities ETF, and Huaxia Robotics ETF, each with inflows exceeding 500 million yuan [3]. - The sectors with the highest net inflows included semiconductors (2.78 billion yuan), securities (1.63 billion yuan), artificial intelligence (1.30 billion yuan), and robotics (1.18 billion yuan) [3][4]. Recent Trends - Over the past five days, ETFs related to securities companies saw inflows exceeding 8.5 billion yuan, while Hong Kong Stock Connect internet-related ETFs attracted over 4.3 billion yuan [4]. - Leading fund companies reported significant inflows in their ETFs, with E Fund's Hang Seng Technology ETF receiving 380 million yuan and its artificial intelligence ETF gaining 190 million yuan [4]. Outflow Summary - On September 23, 18 stock ETFs experienced outflows exceeding 1 billion yuan, with the CSI 300 Index, ChiNext Index, and CSI 500 Index among the hardest hit [6][8]. - The top three ETFs by net outflow were the CSI 300 ETF (1.33 billion yuan), ChiNext ETF (580 million yuan), and CSI 500 ETF (443 million yuan) [6][8].

Core Insights - The number of stock ETFs with assets exceeding 10 billion yuan has increased to 56 as of September 19, 2023, up from 47 at the end of June, indicating a growing interest in these investment vehicles [2][3] - The recent entrants into the "billion club" are primarily industry-themed ETFs, particularly in sectors such as chemicals, resources, robotics, and batteries, with some products experiencing over a tenfold increase in scale since June [3][4] - There has been a significant net inflow of funds into industry-themed ETFs, with 17 ETFs attracting over 1.5 billion yuan in net inflows from September 1 to September 19, 2023, highlighting a trend of capital concentration in specific sectors [5][6] Industry Trends - The rapid growth of specific industry-themed ETFs reflects investor optimism towards certain sectors, driven by economic structural transformation and supportive industrial policies, particularly in high-tech and advanced manufacturing [4][6] - Fund companies have been actively launching and promoting ETFs focused on niche industries, which has contributed to the increase in ETF sizes, aligning with market investment hotspots [4][6] Investor Behavior - Funds flowing into industry-themed ETFs can be categorized into three types: those seeking stable returns (favoring sectors like beverages), those optimistic about industry prospects (investing in robotics), and those attracted by valuation advantages and event-driven opportunities (focusing on brokers, chemicals, and gold stocks) [6][7] - The influx of funds into these ETFs indicates a shift towards a more strategic approach among investors, with some focusing on long-term growth trends while others engage in short-term trading based on market sentiment [7][8] Market Volatility - The volatility of popular ETFs is evident, with significant price fluctuations observed in the leading ETFs during the period from September 1 to September 19, 2023, where some ETFs experienced declines after previous gains [8][9] - Investors are advised to avoid blindly following trends in ETF investments, as the concentration of capital in popular sectors can lead to inflated valuations and potential corrections if market sentiment shifts [9]

Market Overview - The A-share market experienced fluctuations last week, with the overall index declining by 0.18%. The ChiNext index led the gains, while the Shanghai Composite Index fell nearly 2% [1] - The market showed a rebound in the first half of the week, but most broad indices retreated on Thursday following the Federal Reserve's interest rate decision [1] Style and Sector Performance - Small-cap stocks outperformed, with the CSI 1000 index rising by 0.21%, compared to a decline of 0.44% for the CSI 300 index. Growth style stocks also performed well, increasing by 1.45% [2] - Among sectors, coal, electric equipment, and electronics saw the highest gains, while the financial sector faced significant declines, particularly in banking, non-ferrous metals, and non-bank financials [3] Trading Activity - Trading activity in the A-share market increased, with an average daily turnover of 25,178 billion yuan, up by 1,914.31 billion yuan from the previous week. On Thursday, turnover exceeded 30 trillion yuan [4] Fund Flows - Last week, the ETF market saw a net inflow of 178.3 billion yuan, with stock ETFs attracting 206.02 billion yuan. However, money market ETFs experienced a slight outflow of 2.57 billion yuan [5] - Notable net inflows were observed in sectors such as securities companies (100.36 billion yuan), Hong Kong Stock Connect internet (53.73 billion yuan), and robotics industry (40.92 billion yuan) [7] - Conversely, significant net outflows were recorded in the STAR 50 index (67.97 billion yuan) and CSI 300 index (34.52 billion yuan) [5][11] ETF Performance - The median weekly return for stock ETFs was 0.03%, with the ChiNext ETFs showing the highest median return of 2.35%. Technology ETFs also performed well, with a median return of 2.17% [14] - The semiconductor ETFs had strong performances, with several funds showing returns above 7% for the week [16] - On the downside, ETFs related to Hong Kong non-bank financials and industrial non-ferrous metals saw declines of 6.70% and 5.38%, respectively [18][20] Upcoming Developments - A second batch of 14 STAR bond ETFs is set to be launched on September 24 [23]

Core Viewpoint - The market showed strong enthusiasm for investment, with significant inflows into stock ETFs, particularly in sectors like AI and new energy, despite a slight decline in the Shanghai Composite Index [1][2]. Fund Flows - Over the past week, the total net inflow into stock ETFs reached over 286 billion yuan, with notable inflows on three consecutive days [1]. - On Friday alone, the total net inflow into stock ETFs was 71.24 billion yuan, with A-share ETFs contributing 38.42 billion yuan [2]. - Industry-themed ETFs, particularly those tracking the securities index and Hong Kong internet index, saw the highest net inflows, totaling 35.02 billion yuan and 31.71 billion yuan respectively [4]. ETF Performance - The securities ETF experienced a net inflow of nearly 48 billion yuan, while the broker ETF saw a net inflow of approximately 23 billion yuan [6]. - The top-performing ETFs included the securities ETF with a net inflow of 47.91 billion yuan and the Hong Kong internet ETF with 45.69 billion yuan [7]. - The robot ETF from E Fund had a net inflow of 4.2 billion yuan, indicating strong interest in the robotics sector [4][5]. Sector Insights - The AI and new energy sectors are driving market performance, with the ChiNext Index rising by 2.34% and the Sci-Tech 50 Index by 1.84% over the week [2]. - The human-shaped robot industry is expected to gain more attention as production schedules become clearer in the coming years [8]. Outflows - In contrast, several ETFs tracking the ChiNext 50 Index and convertible bond indices faced significant outflows, indicating profit-taking behavior among investors [9].

Core Viewpoint - The stock market has seen significant inflows into ETFs, with a total net inflow exceeding 28.6 billion yuan over the past week, driven by strong market sentiment and sector performance, particularly in AI and new energy sectors [1][2]. Fund Flows - The total net inflow into stock ETFs last week was 71.24 billion yuan, with A-share stock ETFs contributing 38.42 billion yuan [3]. - Industry-themed ETFs and Hong Kong market ETFs led the inflows, with net inflows of 35.02 billion yuan and 31.71 billion yuan, respectively [5]. - The broad-based ETFs experienced a decrease in scale, with a reduction of 47.23 billion yuan [5]. Sector Performance - The securities ETF saw a net inflow of nearly 48 billion yuan, with the brokerage ETF contributing close to 23 billion yuan [7]. - The robot ETF from E Fund had a net inflow of over 20 billion yuan last week, indicating strong interest in the robotics sector [7]. - The Hong Kong stock ETFs, particularly the Hong Kong Internet ETF and the Hong Kong Technology 30 ETF, saw significant inflows of 45.69 billion yuan and 18.43 billion yuan, respectively [8]. Specific ETF Insights - The ETF tracking the CSI A500 index had the highest single-day net inflow of 13.8 billion yuan on September 19, while the ETF tracking the Sci-Tech 50 index faced a net outflow of 15.46 billion yuan [5]. - E Fund's ETFs have shown substantial growth, with the total scale reaching 787.66 billion yuan, an increase of 187.01 billion yuan since 2025 [5]. - The human-shaped robot industry is expected to gain more attention as production schedules become clearer in the next 1-2 years, according to E Fund's fund manager [8].

Core Viewpoint - The stock market showed strong inflows into ETFs, with a total net inflow exceeding 28.6 billion yuan over the past week, indicating a bullish sentiment among investors [2][4]. Fund Flows - The Shanghai Composite Index fluctuated around 3900 points, with significant inflows into stock ETFs over three consecutive days, totaling 286 billion yuan, except for a slight outflow on Tuesday [2][4]. - On Friday alone, the total net inflow into stock ETFs reached 71.24 billion yuan, with A-share ETFs contributing 38.42 billion yuan [4][6]. Sector Performance - Industry-themed ETFs and Hong Kong market ETFs led the inflows, with net inflows of 35.02 billion yuan and 31.71 billion yuan, respectively, while commodity ETFs experienced a net outflow of 9.03 billion yuan [6]. - The A500 index tracking ETF saw a significant single-day net inflow of 13.8 billion yuan on September 19, while the STAR 50 index ETF faced a net outflow of 15.46 billion yuan [7]. Large Fund Companies - Major fund companies like E Fund and Huaxia Fund reported continued net inflows into their ETFs, with E Fund's total ETF scale reaching 787.66 billion yuan, an increase of 187.01 billion yuan since 2025 [9]. - E Fund's Robot ETF and Hong Kong Technology ETF saw net inflows of 4.2 billion yuan and 3 billion yuan, respectively [9]. Specific ETF Performance - The Securities ETF attracted nearly 4.8 billion yuan in net inflows, while the Broker ETF saw inflows of approximately 2.3 billion yuan [11]. - The Hong Kong Internet ETF and the Hong Kong Technology 30 ETF also reported substantial inflows of 45.69 billion yuan and 18.43 billion yuan, respectively [13]. Outflows - In contrast, several ETFs tracking the STAR 50 index, convertible bond index, and CSI 500 index experienced significant outflows, indicating profit-taking behavior among investors [14].

Quantitative Models and Construction Methods 1. Model Name: Diffusion Index Industry Rotation Model - **Model Construction Idea**: The model is based on the principle of price momentum, aiming to capture upward trends in industries through a diffusion index[26][27] - **Model Construction Process**: 1. Calculate the diffusion index for each industry based on price momentum 2. Rank industries by their diffusion index values 3. Select top industries for allocation based on their rankings 4. Adjust the portfolio monthly or weekly based on updated diffusion index rankings[26][27] - **Model Evaluation**: The model has shown stable performance in certain years (e.g., 2022 with an annual excess return of 6.12%) but struggled during market reversals or concentrated market themes, such as in 2024 and 2025[26][33] 2. Model Name: GRU Factor Industry Rotation Model - **Model Construction Idea**: This model leverages GRU (Gated Recurrent Unit) deep learning networks to process high-frequency volume and price data, aiming to identify industry rotation opportunities[38] - **Model Construction Process**: 1. Input high-frequency volume and price data into the GRU network 2. Train the GRU model on historical data to identify patterns in industry rotation 3. Generate factor scores for industries based on the GRU model's output 4. Rank industries by their GRU factor scores and allocate to top-ranked industries[38][34] - **Model Evaluation**: The model performs well in short cycles but struggles in long cycles or extreme market conditions. It has shown difficulty in capturing excess returns in concentrated market themes during 2025[33][38] --- Model Backtesting Results 1. Diffusion Index Industry Rotation Model - **Weekly Average Return**: -1.74%[30] - **Excess Return (Weekly)**: -1.41%[30] - **Excess Return (September 2025)**: -1.88%[30] - **Excess Return (2025 YTD)**: 2.76%[25][30] 2. GRU Factor Industry Rotation Model - **Weekly Average Return**: -0.72%[36] - **Excess Return (Weekly)**: -0.38%[36] - **Excess Return (September 2025)**: -0.10%[36] - **Excess Return (2025 YTD)**: -7.78%[33][36] --- Quantitative Factors and Construction Methods 1. Factor Name: Diffusion Index - **Factor Construction Idea**: Measures the breadth of price momentum across industries to identify upward trends[26][27] - **Factor Construction Process**: 1. Calculate the proportion of stocks in an industry with positive price momentum 2. Aggregate these proportions to derive the diffusion index for the industry 3. Rank industries based on their diffusion index values[27][28] - **Factor Evaluation**: Effective in capturing upward trends but vulnerable to reversals and underperformance in counter-trend markets[26][33] 2. Factor Name: GRU Factor - **Factor Construction Idea**: Utilizes GRU deep learning to analyze high-frequency trading data and generate predictive scores for industry rotation[38] - **Factor Construction Process**: 1. Input high-frequency trading data into the GRU network 2. Train the model to recognize patterns in industry rotation 3. Output factor scores for industries based on the model's predictions[38][34] - **Factor Evaluation**: Strong in short-term predictions but less effective in long-term or extreme market conditions[33][38] --- Factor Backtesting Results 1. Diffusion Index - **Top Industries (Weekly)**: Non-ferrous Metals (0.978), Banking (0.968), Communication (0.946), Electronics (0.877), Automotive (0.874), Retail (0.873)[27] - **Bottom Industries (Weekly)**: Food & Beverage (0.354), Real Estate (0.46), Coal (0.487), Transportation (0.543), Construction (0.574), Building Materials (0.618)[27] 2. GRU Factor - **Top Industries (Weekly)**: Non-ferrous Metals (7.4), Petrochemicals (5.38), Coal (4.17), Steel (4.15), Building Materials (3.46), Non-banking Financials (3.08)[34] - **Bottom Industries (Weekly)**: Comprehensive Finance (-19.42), Utilities (-13.41), Electronics (-13.18), Pharmaceuticals (-11.14), Automotive (-10.07), Consumer Services (-10.04)[34]

Market Overview - The stock indices showed mixed performance this week, with a total net inflow of approximately 25.3 billion yuan into stock ETFs and cross-border ETFs in the Shanghai and Shenzhen markets [1][3]. ETF Performance - Sector-themed ETFs such as those related to securities and robotics attracted significant capital, while chip-related ETFs faced substantial sell-offs [2][15]. - The total trading volume for the Shanghai and Shenzhen markets reached 12.45 trillion yuan, with the Shanghai Composite Index closing at 3820.09 points, down 1.3%, and the Shenzhen Component Index closing at 13070.86 points, up 1.14% [3]. Specific ETF Insights - The large-cap wide-based ETFs experienced a net outflow of 120.05 billion yuan, with the Kweichow Moutai 50 ETF seeing a net outflow of 43.18 billion yuan [7][10]. - The Kweichow Moutai 50 ETF's share count fell below 50 billion, with a year-to-date reduction of 39.528 billion shares and a net outflow of over 40 billion yuan [10]. Sector Analysis - In the sector-themed ETF space, 64 funds saw net inflows exceeding 100 million yuan, with securities ETFs, robotics ETFs, and brokerage ETFs leading the way with net inflows of 4.791 billion yuan, 2.758 billion yuan, and 2.289 billion yuan, respectively [13][14]. - Notably, the securities sector is viewed as having significant upside potential due to its relatively low year-to-date performance, suggesting a favorable entry point for investors [17][18]. Future Outlook - The market is expected to continue its upward trend, supported by easing monetary policy from the Federal Reserve and ongoing domestic growth initiatives [24]. - Upcoming listings of multiple ETFs tracking technology innovation bonds and other sectors are anticipated to attract additional capital [26].