Investment Rating - The industry investment rating is "Overweight" [5] Core Insights - The global computing power construction is entering a new acceleration phase driven by expanding inference demand, innovative financing models, and increased capital expenditure from cloud service providers [1][4][21] - In the AI era, inference demand is driven by three variables: user count, usage duration, and usage depth, which are all expanding simultaneously [2][8][13] - Supply-side cooperation is deepening, with major cloud providers increasing their capital expenditures significantly, indicating a robust growth outlook for computing power infrastructure [3][20] Summary by Sections Demand Side - Inference demand is expanding due to increased user count, with OpenAI's monthly active users surpassing 800 million, indicating rapid penetration of AI applications [2][8] - The depth of usage is evolving as AI products transition from general Q&A tools to complex reasoning models, significantly increasing the frequency of inference calls [9][13] - Usage duration is extending as AI moves from passive response to proactive operation, with products like OpenAI Pulse enabling continuous reasoning, thus enhancing computing power utilization [13][15] Supply Side - Major collaborations are accelerating the pace of data center construction, with OpenAI partnering with leading computing power manufacturers like NVIDIA and AMD for large-scale projects [3][17] - Capital expenditure from North American cloud providers reached $113.4 billion in Q3 2025, reflecting a year-on-year increase of 78.42% and a quarter-on-quarter increase of 19.31% [20][23] - New financing models such as "computing power leasing" and "computing power debt financing" are emerging, alleviating the mismatch between large upfront investments and long-term returns, thus accelerating computing power cluster construction [4][19] Overall Outlook - The combination of sustained growth in inference demand and accelerated construction of computing power clusters is expected to lead the global computing power industry into a new expansion cycle, with a positive outlook for industry prosperity [21]

Investment Rating - The report maintains an "Overweight" rating for the technology and computer sectors [7]. Core Insights - The AI industry is entering a new paradigm characterized by the Scaling Law 2.0, where synthetic data expands the training data ceiling, and the Mid Training paradigm reshapes model evolution paths [2][3]. - The commercial application of AI is transitioning into a scaling phase, with the integration of agent capabilities and transaction loops accelerating industry implementation [2][6]. Summary by Sections Models - Computing power expansion remains the core growth engine, with representative model training computing power expected to grow at an annual rate of 4-5 times from 2010 to 2024, with leading models achieving up to 9 times [3][13]. - The cost of complete training for frontier models is projected to reach the billion-dollar level by 2027 [3][13]. Training - The Mid Training paradigm expands training boundaries by integrating reinforcement learning (RL) into the middle stage, enhancing data generation and optimal allocation [4][16]. - This approach significantly increases data utilization efficiency and is expected to break traditional performance limits [4][16]. Agents - GPT-5 establishes a "unified system" direction, promoting standardization in agent architecture through adaptive collaboration between fast and deep thinking [5][19]. - The real-time router dynamically allocates computing resources based on task complexity, enhancing response efficiency and stability in complex scenarios [5][19]. Applications - The integration of agent capabilities into commercial transactions marks a new phase of AI applications, with OpenAI's Agentic Commerce Protocol enabling AI agents to execute purchases directly [6][22]. - The global AI application landscape is evolving through three stages: productization in 2023, commercialization trials in 2024, and scaling implementation in 2025 [25][26]. - Domestic AI applications are accelerating, with significant advancements in commercial capabilities following the release of models like DeepSeek-R1 [26].

Investment Rating - The communication industry is rated as "Overweight" [10] - The communication equipment manufacturing sector is also rated as "Overweight" [10] Core Insights - The communication sector saw a significant increase in holdings, with the top ten funds holding 7.26% of the communication sector, up 3.58 percentage points from the previous quarter, marking the highest quarterly holding ratio since 2019 [2][3] - The sector's overweight ratio increased to 3.36%, up 2.90 percentage points from the previous quarter [3][14] - Key areas of investment include optical modules, liquid cooling, optical chips, and optical fiber cables, while operators, data centers, and Beidou systems saw reduced holdings [3][14] - The TTM price-to-earnings ratio for the communication sector as of November 7, 2025, is 43.62x, which is at the 69.3% historical percentile since early 2011 [3][14] Summary by Sections Market Performance - The communication index rose by 0.92% last week, while the Shanghai Composite Index increased by 1.08% and the Shenzhen Component Index by 0.19% [2][13] Fund Holdings - In Q3 2025, public funds increased their holdings in the communication sector, with significant increases in companies like Zhongji Xuchuang, Xinyi Sheng, Tianfu Communication, Yingweike, and Yuanjie Technology, with total market value increases of 82.43 billion, 78.93 billion, 9.29 billion, 4.06 billion, and 3.79 billion respectively [3][25] - The top five companies by increase in holdings during Q3 2025 saw stock price increases of 176.8%, 188.0%, 110.8%, 169.2%, and 120.3% respectively [25][26] Key Companies and Recommendations - Recommended companies include Xinyi Sheng, Zhongxing Communication, Ruijie Network, China Mobile, China Telecom, China Unicom, and Hengtong Optoelectronics, all rated as "Buy" [10][58] - The report highlights the potential for AI computing chain investments, particularly in optical modules and related technologies [2][3] Valuation and Future Outlook - The report anticipates continued growth in the AI computing sector, with companies like Xinyi Sheng expected to benefit significantly from the demand for high-speed products [59][62] - Adjustments to profit forecasts for companies like Zhongxing Communication and Ruijie Network reflect a cautious outlook due to market conditions, but long-term growth potential remains [59][62]

Government Shutdown Duration and Impact - As of November 9, the U.S. government has been shut down for 40 days, marking a historical record[2] - The shutdown is primarily due to political polarization, with both parties believing it benefits them[2] - It is expected that the government will remain closed for at least another 1-2 weeks, likely reopening before Thanksgiving (November 27)[3] Economic and Employment Effects - The shutdown has resulted in at least 670,000 federal employees being furloughed and approximately 730,000 working without pay[10] - If the shutdown continues until December 1, the total unpaid wages could reach approximately $21 billion[10] - The impact on GDP growth for Q4 2025 is projected to be a reduction of over 1 percentage point, with a rebound expected in Q1 2026[5] Data Release Delays - Key economic data for September and October, including non-farm payrolls and CPI, have been delayed due to the shutdown[4] - If the government reopens in 1-2 weeks, some data may be released shortly thereafter, but the timing remains uncertain[19] - October non-farm payroll data may be published alongside November data in early December[20] Market and Policy Implications - The shutdown has led to a rise in the Treasury General Account (TGA) balance by $62.7 billion, which may tighten liquidity marginally[36] - The Federal Reserve is expected to lower interest rates once in December 2025 and potentially 1-2 more times in June 2026[37] - The reopening of the government is anticipated to alleviate some liquidity pressures in the market[36]

Investment Rating - The investment rating for the company is maintained as "Buy" with a target price of RMB 31.50 [1][5]. Core Views - The report indicates that the demand for oxalic acid is expected to improve due to growth in the rare earth and lithium battery sectors, with a competitive landscape that is favorable [1]. - The company has a production capacity of 700,000 tons per year for oxalic acid and 300,000 tons per year for caprolactam, positioning it to benefit from the anticipated recovery in market conditions [1]. - The report highlights that the domestic competition for oxalic acid is relatively good, with the company holding a significant market share [3]. Summary by Sections Oxalic Acid Market Outlook - The apparent consumption of oxalic acid in China is projected to reach approximately 417,000 tons in 2024, representing a year-on-year increase of 87%. The rare earth separation and new energy sectors contribute significantly to this demand [2]. - The oxalic acid precipitation method is a mature technology for rare earth separation, which is expected to support continued demand growth as the strategic importance of rare earth resources increases [2]. Competitive Landscape - As of November 2025, the domestic oxalic acid production capacity is about 1.19 million tons, with a concentrated market where the company holds a 59% share [3]. - The company utilizes a gasification platform to co-produce dimethyl carbonate and oxalic acid, which offers cost and environmental advantages over traditional methods [3]. Caprolactam Market Dynamics - The caprolactam industry is undergoing a "de-involution" process, with companies discussing measures to reduce production by 20% and increase prices by RMB 100 per ton [4]. - The report anticipates that the supply-demand balance will improve, aiding in price recovery for various chemical products [4]. Financial Projections - The company is expected to achieve net profits of RMB 32 billion, RMB 45 billion, and RMB 54 billion for the years 2025 to 2027, respectively, with corresponding EPS of RMB 1.50, RMB 2.10, and RMB 2.56 [5][9]. - The target price of RMB 31.50 reflects a 15x PE ratio for 2026, considering the company's scale advantages and growth potential from new projects [5].

Macro Insights - Core inflation in China rebounded in October 2025, with CPI rising 0.2% year-on-year compared to a previous decline of 0.3%, exceeding Bloomberg's consensus expectation of -0.1% [2] - The global manufacturing PMI showed a slight recovery in October, remaining above the threshold for the third consecutive month, indicating resilience in the global manufacturing cycle despite a marginal decline in new export orders [2][3] - The U.S. employment data showed mixed results, with ADP employment figures exceeding expectations, while uncertainty around tariff policies increased due to potential legal challenges [3] Industrial Production and Exports - High-frequency indicators from ports showed a slight year-on-year recovery, suggesting a marginal improvement in export sentiment for November, while industrial production remained resilient [4] - October exports experienced a short-term disturbance, declining to -1.1% year-on-year from a previous 8.3% [4] Investment Strategy - The A-share market showed a rebound, led by manufacturing and cyclical sectors, while technology stocks continued to adjust [5] - The report suggests a shift in focus towards profit-driven investment strategies, with an emphasis on advanced manufacturing and consumer sectors as signs of economic recovery emerge [5] - A "barbell" strategy is recommended, focusing on dividend stocks and small-cap stocks, while maintaining a balanced sector allocation [7] Commodity Market - Commodity strategies showed a slight increase of 0.57% over the past two weeks, with a year-to-date increase of 3.17% [8] - The report highlights a strong performance in the commodity term structure simulation, particularly in agricultural products and industrial metals, while energy and chemical sectors faced declines [8] Fixed Income Market - The fixed income market is experiencing a tug-of-war at key levels, with expectations of a narrow trading range in the short term due to concerns over AI bubbles and year-end profit-taking [13] - The report notes a seasonal effect in the bond market, with expectations for a year-end rally despite pressures from rising interest rates and market dynamics [14] Transportation Sector - The transportation sector showed strong profitability in Q3 2025, with airports, oil transportation, and railways performing well, while logistics and express delivery exhibited mixed results [17] - The report recommends specific stocks in the transportation sector that are expected to benefit from ongoing profitability improvements [17] Energy and New Power - The recent white paper on carbon neutrality emphasizes the importance of new energy storage and grid upgrades, indicating a favorable outlook for related sectors [18] - The report identifies three key areas for investment: new energy and storage, grid upgrades, and traditional power sources [18] Consumer Sector - The consumer sector is undergoing transformation, with opportunities arising from supply innovations and changing consumer preferences [19] - The report highlights trends in various consumer segments, including beauty and lifestyle products, indicating potential for growth in these areas [19] Chemical Industry - The chemical industry is expected to see a recovery in product demand due to supply-side reforms and improved pricing dynamics [27] - Specific companies within the chemical sector are recommended based on their potential to benefit from these trends [27][30]

Quantitative Models and Construction Methods 1. **Model Name: Commodity Term Structure Simulation Portfolio** - **Model Construction Idea**: The model is based on the roll yield factor to depict the contango and backwardation states of commodities, dynamically going long on commodities with high roll yields and shorting those with low roll yields[23][24] - **Model Construction Process**: - Identify the roll yield factor to determine the contango and backwardation states of commodities - Dynamically adjust the portfolio to go long on commodities with high roll yields and short on those with low roll yields - The portfolio's latest holdings are mainly long agricultural products and industrial metals, and short energy and chemical products[23][24][30] - **Model Evaluation**: The model has shown strong performance, achieving a new high since 2010[23][28] 2. **Model Name: Commodity Time Series Momentum Simulation Portfolio** - **Model Construction Idea**: The model uses multiple technical indicators to depict the medium and long-term trends of domestic commodities, dynamically going long on assets with upward trends and shorting those with downward trends[23][24] - **Model Construction Process**: - Use technical indicators to identify medium and long-term trends in commodities - Dynamically adjust the portfolio to go long on assets with upward trends and short on those with downward trends - The latest holdings are mainly long industrial metals and short energy and agricultural products[33][36] - **Model Evaluation**: The model has shown mixed performance, with some recent gains but an overall decline for the year[33][34] 3. **Model Name: Commodity Cross-Sectional Inventory Simulation Portfolio** - **Model Construction Idea**: The model is based on the inventory factor to depict changes in the fundamentals of domestic commodities, dynamically going long on assets with decreasing inventories and shorting those with increasing inventories[23][24] - **Model Construction Process**: - Identify the inventory factor to determine changes in commodity fundamentals - Dynamically adjust the portfolio to go long on assets with decreasing inventories and short on those with increasing inventories - The latest holdings are mainly long industrial metals and agricultural products, and short energy and chemical products[37][40][42] - **Model Evaluation**: The model has shown strong performance, achieving a new high since 2010[37][38] Model Backtesting Results 1. **Commodity Term Structure Simulation Portfolio** - **Recent Two-Week Return**: 0.90%[26] - **Year-to-Date Return**: 6.44%[28] - **Top Contributors**: Methanol (0.34%), Palm Oil (0.22%), Polypropylene (0.14%)[30] - **Top Detractors**: Copper (-0.09%), Rapeseed Oil (-0.10%), Soybean Meal (-0.20%)[30] 2. **Commodity Time Series Momentum Simulation Portfolio** - **Recent Two-Week Return**: 0.47%[26] - **Year-to-Date Return**: -2.66%[33] - **Top Contributors**: Methanol (0.29%), Aluminum (0.21%), Palm Oil (0.16%)[33] - **Top Detractors**: PTA (-0.06%), Copper (-0.08%), Hot Rolled Coil (-0.09%)[33] 3. **Commodity Cross-Sectional Inventory Simulation Portfolio** - **Recent Two-Week Return**: 0.34%[26] - **Year-to-Date Return**: 5.90%[37] - **Top Contributors**: Aluminum (0.30%), Polypropylene (0.29%), Polyethylene (0.28%)[40] - **Top Detractors**: Rubber (-0.18%), Zinc (-0.24%), Methanol (-0.35%)[40]

- The "A-Share Market Timing Model" evaluates the overall directional judgment of the A-share market using four dimensions: valuation, sentiment, funds, and technicals. The model generates daily signals with values of 0, ±1, representing neutral, bullish, and bearish views, respectively. The model's logic includes mean reversion for valuation and sentiment, and trend continuation for funds and technicals[2][9][15] - The "Style Timing Model" favors a barbell structure of dividend and small-cap styles. For the dividend style, the model uses the relative momentum of the CSI Dividend Index to the CSI All Share Index, the 10Y-1Y term spread, and the interbank pledged repo transaction volume. For the small-cap style, the model employs a trend model based on the difference in momentum and trading volume between small-cap and large-cap stocks[3][17][21] - The "Industry Rotation Model" uses genetic programming to directly extract factors from the volume, price, and valuation characteristics of industry indices. The model updates its factor library quarterly and rebalances weekly, selecting the top five industries with the highest multi-factor composite scores for equal-weight allocation[4][29][34] - The "China Domestic All-Weather Enhanced Portfolio" employs a macro factor risk parity framework, selecting four macro risk sources: growth above/below expectations and inflation above/below expectations. The model actively overweights favored quadrants based on macro expectation momentum, adjusting monthly[5][39][42] - The "A-Share Market Timing Model" achieved a year-to-date return of 36.03%, with an excess return of 8.86% over the Wind All A Index, which had a return of 27.18%[2][9] - The "Style Timing Model" for the dividend style yielded a year-to-date return of 25.04%, with an excess return of 7.83% over the benchmark, which had a return of 17.21%[17][20] - The "Style Timing Model" for the small-cap style achieved a year-to-date return of 78.29%, with an excess return of 30.25% over the benchmark, which had a return of 48.04%[22][27] - The "Industry Rotation Model" achieved a year-to-date return of 40.67%, outperforming the industry equal-weight benchmark by 17.96 percentage points[4][32] - The "China Domestic All-Weather Enhanced Portfolio" achieved a year-to-date return of 11.10%, with a Sharpe ratio of 2.22, a maximum drawdown of 2.67%, and a Calmar ratio of 5.15[5][40][43]

Group 1 - The report highlights the importance of balanced asset allocation amid market volatility, particularly in the context of recent adjustments in technology stocks and increased global market volatility [2][3]. - Major indices such as the Nikkei 225, KOSPI, NASDAQ, and Hang Seng Tech experienced declines of 4.1%, 3.7%, 3.0%, and 1.2% respectively, with significant pullbacks in large AI-related companies [2][3]. - The report suggests that the current technology market is approaching a critical juncture similar to 1998, recommending a more balanced sector allocation for 2026 [3][4]. Group 2 - The report identifies two main catalysts for the recent market fluctuations: disruptions in AI narratives and liquidity pressures, with the former being the dominant factor [3][4]. - As of Q3 2025, the capital expenditure to revenue ratio for the Magnificent 7 (Mag7) reached 17.4%, marking a new high since 2023 and surpassing levels seen during the ".com" boom [3][4]. - The liquidity environment remains tight, with interbank liquidity pressures persisting despite expectations of interest rate cuts, as indicated by rising TGA account balances and a decline in bank reserves [4][5]. Group 3 - The report emphasizes the need for risk management strategies, advising investors to reduce long positions and consider hedging through derivatives or increasing exposure to VIX [5]. - Despite the current market challenges, the report maintains that there is still medium-term investment value in Hong Kong technology stocks, particularly in the context of China's AI industry development [5][6]. - The report notes that the valuation of Chinese stock indices remains attractive compared to global peers, supported by the "14th Five-Year Plan" which emphasizes technological self-sufficiency [5][6].

Investment Rating - The report maintains a "Buy" rating for DoorDash with a target price of $280.60 [6]. Core Insights - DoorDash reported a total gross order value (GOV) of $25.02 billion for Q3 2025, a year-over-year increase of 25.1%, exceeding consensus expectations by 1.9% [1]. - Revenue reached $3.45 billion, up 27.3% year-over-year, also surpassing expectations by 2.6%, driven by significant growth in monthly active users and improvements in average order value and frequency [1][2]. - GAAP net profit was $240 million, below the expected $290 million, while adjusted EBITDA was $750 million, a 42% year-over-year increase, aligning closely with expectations [1][3]. - The company anticipates a short-term profit pressure due to rising operational costs and planned reinvestments of several hundred million dollars in 2026, which are expected to benefit long-term growth [1][3]. Summary by Sections Performance Overview - In Q3 2025, DoorDash's order volume reached 776 million, a 21% year-over-year increase, exceeding expectations by 0.4% [2]. - The average order value (AOV) also surpassed expectations, contributing to the overall GOV exceeding forecasts [2]. Growth Drivers - Growth is driven by a resurgence in core categories, particularly in the U.S. restaurant sector, where GOV growth reached a three-year high [2]. - The international business is expanding robustly, with growth rates in several countries outpacing local competitors, and the integration of Deliveroo is expected to enhance scale and synergy [2]. Cost and Profitability - The increase in operational costs led to a GAAP net profit of $240 million, which was lower than the anticipated $290 million [3]. - The company’s expense ratio for Q3 2025 was 38.6%, higher than the expected 38.0%, primarily due to seasonal increases in rider costs and a higher proportion of low-margin international business [3]. Financial Projections - Revenue forecasts for 2025-2027 are adjusted to $13.76 billion, $17.34 billion, and $21.25 billion, reflecting increases of 4%, 8%, and 12% respectively [4]. - Adjusted EBITDA projections for the same period are revised to $2.8 billion, $3.7 billion, and $4.7 billion, with slight downward adjustments due to lower-than-expected Q4 guidance [4]. Valuation - The report suggests a valuation premium for DoorDash, switching to a 2026 valuation based on a 30x EV/EBITDA multiple, resulting in a target price of $280.60 [4].