Investment Rating - The report does not explicitly state an investment rating for the nuclear power industry, but it suggests a positive outlook based on the revival and growth potential of the sector. Core Insights - The global nuclear power industry is entering a new development opportunity driven by energy security, low-carbon transition, and the demand for AI computing power. Nuclear power is being reassessed for its strategic value as a stable and efficient clean baseload energy source [3][5]. - As of the end of 2024, there are 417 operational nuclear reactors across 31 countries, with a total installed capacity of 377 GW. The International Atomic Energy Agency (IAEA) forecasts that global nuclear capacity will reach between 561 GW (low estimate) and 992 GW (high estimate) by 2050, representing increases of 48.8% and 163.1% respectively compared to 2024 [3][5][6]. - The latest technological focus in nuclear power is on fourth-generation nuclear systems and small modular reactors (SMRs), with significant advancements being made in these areas [3][62]. Summary by Sections 1. Nuclear Power Revival and Demand - The nuclear power industry is experiencing a revival due to urgent climate change needs, energy security considerations, and the explosive growth of AI data centers, which significantly increase electricity demand [5][16]. - The IAEA predicts that by 2050, global nuclear capacity could reach between 561 GW and 992 GW, marking substantial growth from 2024 levels [5][6]. 2. Nuclear Power vs Other Energy Sources - Nuclear power is recognized for its ability to provide stable, continuous electricity, making it a crucial baseload power source in the transition to a cleaner energy system [37][42]. - Compared to coal, nuclear power has a significantly lower carbon footprint, with lifecycle emissions being only about 1% of coal's [42][43]. 3. Nuclear Power Technology Updates - The report highlights the evolution of nuclear technology through four generations, with current advancements focusing on third-generation deployment and fourth-generation demonstration breakthroughs [3][62]. - SMRs are gaining attention for their safety, modular construction, and lower costs, making them attractive to major tech companies [48][49]. 4. Nuclear Power Industry Chain - The report suggests focusing on stable revenue operators in the nuclear sector, such as CEG and China General Nuclear Power, while also identifying high-growth opportunities in companies involved with SMRs and nuclear fuel [3][22]. - The nuclear power sector is expected to see significant investment, with estimates of around $74.5 billion in orders related to nuclear power projects in the U.S. alone [22][23]. 5. Investment Logic - The report emphasizes the importance of nuclear power in meeting the growing electricity demand from AI data centers and the strategic shift towards energy independence in various countries [22][26]. - Countries are increasingly recognizing the need for resilient power systems, with policies being enacted to support nuclear power development [26][27].

Investment Rating - The report maintains an "Outperform" rating for the machinery equipment industry as of March 1, 2026 [1]. Core Insights - The robotics index underperformed the market, with the Zhongzheng Kexin Robotics Index dropping by 4.3%, lagging behind the Shanghai Composite Index by 3.8 percentage points and the CSI 300 Index by 4.3 percentage points [4][11]. - BMW plans to deploy humanoid robots named AEON at its Leipzig factory in Europe, which will begin large-scale pilot testing in summer 2026. These robots are expected to enhance production efficiency, having previously supported the production of 30,000 BMW X3 vehicles in the U.S. [15][16]. - Yushu Technology launched its fourth robotic dog, Unitree As2, which boasts performance capabilities twice that of its predecessor, Go2, with a peak torque of 90N.m and a maximum speed of 5m/s. The company anticipates that 80% of its robotic dogs will be used in research, education, and consumer sectors by 2024 [18][19]. - Qianxun Intelligent, a leading company in embodied intelligence, completed two rounds of financing totaling nearly 2 billion yuan, with plans to invest in developing large-scale models and expanding technology applications across various sectors [19]. Summary by Sections Market Review - The robotics index experienced a decline of 4.3% from February 23 to March 1, 2026, underperforming compared to major indices [4][11]. Industry Dynamics - BMW's deployment of humanoid robots in its Leipzig factory marks a significant step in integrating AI-based robotics into its European production system, leveraging successful experiences from its U.S. operations [15][16]. - Yushu Technology's new robotic dog, Unitree As2, features enhanced performance metrics and aims to capture a significant share of the global market for quadruped robots [18]. - Qianxun Intelligent's recent funding rounds reflect strong investor confidence and a strategic focus on scaling technology across multiple applications [19].

Domestic Economic Developments - The LPR remains unchanged for the ninth consecutive month, with the 1-year LPR at 3.0% and the 5-year LPR at 3.5%[8] - The government emphasizes "releasing silver-haired consumption demand" as a key strategy to address aging population challenges, indicating a shift towards viewing the elderly as a vital consumer group[10] - Shanghai's new real estate policy aims to stabilize the market by relaxing purchase restrictions and enhancing loan limits, with the maximum loan for first-time buyers increased from 1.6 million to 2.4 million yuan[12] International Economic Trends - The offshore and onshore RMB exchange rates reached new highs since April 2023, driven by a weak USD and increased corporate demand for currency exchange[20] - Japan's CPI for February shows a year-on-year increase of 1.6%, falling below the Bank of Japan's 2% target, primarily due to government subsidies and a slowdown in food cost growth[22] Market Data Insights - Brent crude oil prices increased by 0.27% week-on-week, while iron ore prices decreased by 1.86%[24] - The average price of domestic polysilicon dropped by 4.65%, and lithium carbonate prices fell by 5.94% week-on-week[24] Policy and Economic Outlook - The central bank is expected to maintain a cautious approach to monetary policy, with potential for targeted rate cuts to support the real estate market and economic stability[9] - The upcoming political meetings are set to finalize the "15th Five-Year Plan," focusing on structural optimization and efficiency improvements in the economy[14]

1. Report Industry Investment Rating - Not provided in the given content 2. Core Viewpoints of the Report - Short - term bond ETFs are still constrained by the stock market. The net inflow of interest - rate bond ETFs, credit - bond ETFs, and convertible - bond ETFs last week was -3.293 billion yuan, -6.21 billion yuan, and +930 million yuan respectively, with a total net inflow of -8.573 billion yuan in the bond ETF market. The strong performance of the A - share market after the Spring Festival and the real - estate policy in Shanghai have affected the bond market sentiment [2][5]. - In the future, the fundamental repair expectations and allocation needs of interest - rate and credit - bond ETFs resonate. Long - end varieties have gaming value before the Two Sessions. The underlying assets of interest - rate and credit - bond ETFs have limited downside space, and the market is expected to continue the trading - style market before the policy implementation of the Two Sessions [2][15]. - For convertible - bond ETFs, the bull market in the equity market drives valuation expansion. Due to policy dividends in 2026, the net value of convertible - bond ETFs may still have room to rise. Given the limitations in individual bond selection, convertible - bond ETFs have higher tool - based allocation cost - effectiveness [2][15]. 3. Summary According to Relevant Catalogs 3.1 Various Bond ETF Fund Inflow Situations - The net inflow of interest - rate bond ETFs, credit - bond ETFs, and convertible - bond ETFs last week was -3.293 billion yuan, -6.21 billion yuan, and +930 million yuan respectively. The total net inflow of the bond ETF market was -8.573 billion yuan, with a cumulative net inflow of 7.818 billion yuan this month and -100.529 billion yuan this year. The bond ETF fund scale was 734.641 billion yuan as of February 27, 2026, down 1.25% from the previous week's close and 11.41% from the beginning of the year, accounting for 13.64% of the total market ETF scale, a decrease of 23bp from the previous weekend [5]. - After the Spring Festival, funds flowed out of short - term financing ETFs. The net inflow of convertible - bond ETFs led last week, while short - term financing ETFs had a net outflow of 2.938 billion yuan, and科创债 ETFs and treasury - bond ETFs also had relatively large net outflows [6]. 3.2 Share and Net Value Trends of Various Bond ETFs and Representative Products - As of February 27, 2026, the shares of treasury - bond, policy - financial - bond, local - bond, benchmark - market - making credit - bond,科创债, corporate - bond, short - term - financing, urban - investment - bond, and convertible - bond ETFs were 646.59 million shares, 350.39 million shares, 159.41 million shares, 1.02384 billion shares, 2.72782 billion shares, 334.91 million shares, 682.84 million shares, 3.05238 billion shares, and 5.50985 billion shares respectively, with changes of -2.7%, -3.5%, +0.3%, -0.4%, -1.1%, +0.2%, -3.7%, 0.0%, and +1.2% compared to February 13, 2026, and changes of +3.5%, +3.8%, +0.9%, -3.0%, -4.4%, +4.8%, +18.9%, +9.2%, and +3.5% compared to the end of last month [22]. - The net value of the 30 - year treasury - bond ETF led the decline among major bond ETFs. As of February 27, 2026, the net values of the 30 - year treasury - bond ETF, policy - financial - bond ETF, 0 - 4 - year local - bond ETF, corporate - bond ETF, short - term - financing ETF Haifutong, urban - investment - bond ETF Haifutong, and convertible - bond ETF Boshi were 1.2244 yuan, 1.1566 yuan, 1.1651 yuan, 1.1960 yuan, 1.1309 yuan, 1.4445 yuan, and 1.4570 yuan respectively, with changes of -0.62%, -0.11%, 0.03%, 0.03%, 0.05%, 0.04%, and -0.23% compared to February 27, 2026, and changes of 0.16%, 0.33%, 0.10%, 0.13%, 0.12%, 0.16%, and 0.80% compared to the end of last month [25]. 3.3 Share and Net Value Trends of Each Benchmark - Market - Making Credit - Bond ETF - The shares of existing credit - bond ETFs fluctuated slightly, with E Fund and Boshi achieving reverse growth in shares. As of February 27, 2026, the shares of 8 credit - bond ETFs were 103.46 million shares, 88.41 million shares, 106.54 million shares, 96.47 million shares, 174.83 million shares, 216.90 million shares, 92.90 million shares, and 144.32 million shares respectively, with changes of 0.00%, 2.31%, -1.84%, 0.00%, -1.13%, 0.60%, -2.11%, and -0.69% compared to February 13, 2026 [29]. - The net value continued to rise, with Dacheng, Southern, and Haifutong leading the increase. As of February 27, 2026, the net values of 8 credit - bond ETFs were 1.0193, 1.0185, 1.0170, 1.0173, 1.0123, 1.0153, 1.0163, and 1.0153 respectively, with changes of 0.03%, 0.04%, 0.03%, 0.04%, 0.03%, 0.02%, 0.04%, and 0.02% compared to February 13, 2026, and changes of 0.21%, 0.19%, 0.20%, 0.17%, 0.21%, 0.18%, 0.21%, and 0.20% compared to the end of last month [31]. 3.4 Share and Net Value Trends of Each 科创债 ETF - The shares of 科创债 ETFs decreased slightly overall, with Yongying achieving reverse capital absorption. The net inflow of shares last week was -31.73 million shares, a decrease of 1.15% from the previous week. As of February 27, 2026, the top - three products in terms of shares were 科创债 ETF Jiashi, 科创债 ETF Yinhua, and 科创债 ETF Penghua, with 221.81 million shares, 197.39 million shares, and 196.81 million shares respectively. 科创债 ETF Yinhua, 科创债 ETF Jiashi, and 科创债 ETF Zhaoshang had the top - three net outflows of shares, with 8.40 million shares, 7.63 million shares, and 7.00 million shares respectively. 科创债 ETF Yongying had the leading net inflow of shares (+4.60 million shares) [34]. - The net value continued to rise. As of February 27, 2026, the top - ranked products in terms of net value among 24 科创债 ETFs were 科创债 ETF Wanjia, 科创 ETF Huatai Bairui, and 科创债 ETF Yongying, with 1.0092, 1.0089, and 1.0088 respectively. The median net values of the first - batch and second - batch 科创债 ETFs last week were 1.0047 and 1.0073 respectively, up 0.03% and 0.02% from the previous week's close. The median net values of products tracking the CSI AAA 科创债, Shanghai AAA 科创债, and Shenzhen AAA 科创债 were 1.0068, 1.0051, and 1.0089 respectively, up 0.03%, 0.03%, and 0.02% from the previous week's close [42]. 3.5 Market Performance of Single Bond ETFs Last Week - The net values of bond ETF products generally declined last week. The 30 - year treasury - bond ETF, 30 - year ETF Boshi, and convertible - bond ETF Haifutong led the decline, down 0.67%, 0.63%, and 0.47% respectively from the previous week. In terms of premium/discount rates, convertible - bond ETF Boshi, 30 - year treasury - bond ETF, and treasury - bond ETF Huaxia had the leading premium rates, at +0.05%, +0.02%, and +0.02% respectively. In terms of scale changes, 科创债 ETF Yongying (+464 million yuan), convertible - bond ETF Boshi (+586 million yuan), and convertible - bond ETF Haifutong (+344 million yuan) had the leading net inflows [45]. 3.6 Marginal Changes in the PCF Lists of Benchmark - Market - Making Credit - Bond and 科创债 ETFs - The estimated modified durations of benchmark - market - making credit - bond ETFs decreased across the board last week. The average modified duration of bonds listed in the PCF list of each bond ETF was used as the estimated modified duration of the ETF. Among products tracking the Shanghai - market - making 科创债 index, the average modified durations of newly added bonds in the PCF lists of corporate - bond ETF Southern and credit - bond ETF Haifutong were 6.10 years and 4.32 years respectively. There were no repeated inclusions or exclusions of bonds in the PCF lists of benchmark - market - making credit - bond ETFs last week [48]. - The estimated modified durations of 科创债 ETFs generally shortened last week, with 科创债 ETF Tianhong and 科创债 ETF Dacheng experiencing relatively large shortenings. The PCF lists of each 科创债 ETF were adjusted slightly last week. The modified durations of newly added bonds were generally between 2 - 5 years. Among the excluded bonds, the average modified durations of bonds excluded from the PCF lists of 科创债 ETF Southern and 科创债 ETF Tianhong, which track the CSI AAA 科创债 index, were significantly longer, at 5.71 years and 7.87 years respectively [51]. - Yongtongshang K1 was repeatedly excluded from the PCF lists of 科创债 ETFs last week due to its upcoming maturity. Bonds such as 26 Shudao K2, 26 Jieneng K1, and 26 Jieneng K2 were included in multiple 科创债 ETFs and are newly listed bonds [53].

1. Report Industry Investment Rating No information about the report industry investment rating is provided in the given content. 2. Core Viewpoints of the Report - The bond market faced callback pressure last week, with the yield of 10-year Treasury bonds fluctuating around the key point of 1.8%, and the market's long - short game intensified. The bond market's yield decline encountered resistance after the Spring Festival, and differences among institutions emerged. The market may present a volatile pattern in the short term [2][85]. - The operation directions and term preferences of trading institutions are still significantly differentiated. Whether small and medium - sized banks and securities firms can form a synergy is an important variable for the market to break the volatile pattern. The securities firms were the core driving force for the yield decline, but their scale of increasing holdings of 7 - 10 - year Treasury bonds decreased last week. Funds sold a large amount of 20 - 30 - year Treasury bonds and 7 - 10 - year policy - financial bonds, while small and medium - sized banks showed a strong willingness to buy ultra - long Treasury bonds and 7 - 10 - year policy - financial bonds [2][86]. - Looking forward, the bond market may show a volatile trend in the short term. Factors such as the potential increase in government bond supply, the seasonal increase in bank loan scale at the end of the quarter, and the lack of synergy among trading institutions make it unlikely for the yield of 10 - year Treasury bonds to break through the previous low. However, the probability may increase due to the rapid deterioration of the geopolitical situation [2][87]. 3. Summary According to the Directory 3.1 Important Matters - In February, the net MLF injection was 300 billion yuan. The central bank carried out a 600 - billion - yuan MLF operation on February 25, with a maturity of 1 year. After deducting the 300 - billion - yuan due MLF, the net injection was 300 billion yuan. As of February 27, the outstanding MLF scale was 7.25 trillion yuan [5]. - On February 28, the US and Israel launched a large - scale joint military strike against Iran, and Iran counterattacked with missiles and drones, leading to a sharp escalation of the situation in the Middle East [8]. 3.2 Money Market 3.2.1 Open Market Operations and Fund Interest Rate Trends - From February 24 to February 28, the central bank injected 164.1 billion yuan through 7 - day reverse repurchase operations, with 225.24 billion yuan due, resulting in a net injection of - 61.14 billion yuan. It is expected that 152.5 billion yuan of base currency will mature and be withdrawn from March 3 to March 6 [10]. - After the Spring Festival, the money market became looser. As of February 28, R001, R007, DR001, and DR007 were 1.340%, 1.507%, 1.319%, and 1.503% respectively, with changes of 1.11BP, 16.27BP, 0.68BP, and 18.23BP compared to February 14. The interest rate centers also changed to some extent [14]. 3.2.2 Certificate of Deposit Interest Rate Trends and Repurchase Transaction Situations - In the primary market, the issuance scale of inter - bank certificates of deposit last week was 45.435 billion yuan, a decrease of 25.857 billion yuan compared to before the festival. The maturity scale was 66.676 billion yuan, a decrease of 28.059 billion yuan compared to before the festival. The net financing scale was - 21.241 billion yuan, an increase of 2.202 billion yuan compared to before the festival. The cumulative issuance scale of inter - bank certificates of deposit in 2025 has reached 3.4 trillion yuan as of the 9th week [20]. - The issuance scale of inter - bank certificates of deposit was the largest for joint - stock banks, with a net financing scale of - 6.096 billion yuan. The issuance interest rates of inter - bank certificates of deposit decreased compared to before the festival [23][25]. - In the secondary market, supported by relatively loose liquidity, the yields of inter - bank certificates of deposit of various terms were generally on a downward trend. The yield of AAA - rated 1 - month inter - bank certificates of deposit decreased by 7.83BP to 1.46%, and the 1Y - 3M spread was at the 47.59% quantile level [29]. 3.3 Bond Market 3.3.1 Primary Market - Last week, the number of interest - rate bond issuances was 46, with an actual issuance amount of 787.42 billion yuan, a maturity amount of 409.832 billion yuan, and a net financing amount of 377.588 billion yuan. The issuance rhythm of national bonds and local bonds in February 2026 was higher than the historical average [31]. - As of February 28, the cumulative net financing scale of various national bonds in 2026 was about 0.83 trillion yuan, and that of various local bonds was about 1.77 trillion yuan, both faster than the average levels in the same period from 2022 to 2025 [31]. - The net supply scale of interest - rate bonds increased last week. The net financing amounts of national bonds, local bonds, and policy - financial bonds were 370 billion yuan, 195.228 billion yuan, and - 187.64 billion yuan respectively [39]. - As of last week, 0.8 trillion yuan of special refinancing bonds had been issued, mainly with long - term and ultra - long - term maturities. The top - ranking regions in terms of issuance scale accounted for about 58.45% of the total issuance scale [40]. 3.3.2 Secondary Market - Last week, the bond market faced callback pressure under the profit - taking of small and medium - sized banks and the stop - loss of securities firms, with long - term and ultra - long - term bonds generally rising. On Friday and Saturday, the interest rates recovered slightly under the buying support of large banks [31][43]. - The yields of 1 - year, 3 - year, 5 - year, 7 - year, 10 - year, and 30 - year Treasury bonds changed by 0.23BP, 0.34BP, 0.59BP, 1.25BP, - 1.46BP, and 2.66BP respectively. The 10Y - 1Y Treasury bond yield spread changed from 47.54BP before the festival to 45.85BP. The yields of the same - term policy - financial bonds also changed, and the implied tax rate of 10 - year policy - financial bonds increased slightly [43]. - The average daily turnover rate of the 10 - year Treasury bond active bond (250016) decreased, while that of the 10 - year policy - financial bond active bond (250220) increased [45]. - The average spread between the 10 - year Treasury bond active bond (250016) and the secondary - active bond (260005) was 0.12BP, and the spread between the 10 - year policy - financial bond active bond (250220) and the secondary - active bond (250215) continued to be maintained. There may be investment opportunities in the spread compression [48]. - The 10 - 1 - year Treasury bond term spread was 45.85BP, and the 30 - 1 - year Treasury bond term spread widened to 95.58BP. The long - term Treasury - local bond spread widened, while the ultra - long - term Treasury - local bond spread narrowed [52][53]. 3.4 Institutional Behavior Tracking - In January 2026, the institutional leverage ratio decreased seasonally, while the leverage ratios of banks and securities firms increased significantly. The leverage ratios of commercial banks, securities firms, and other institutions in the inter - bank market in January 2026 were about 111.11%, 191.81%, and 132.51% respectively [58][59]. - The 20 - day moving average of the daily trading volume of inter - bank pledged repurchase last week was 7.71 trillion yuan, a decrease of about 0.56 trillion yuan compared to before the festival. The daily leverage trading volume decreased as the market callback [62]. - In the cash bond market, large banks' increasing holdings of 5 - 10 - year Treasury bonds continued to decline, while their increasing holdings of Treasury bonds within 5 years recovered. Small and medium - sized banks continued to take profits on Treasury bonds within 10 years and increased their positions in Treasury bonds over 10 years and 5 - 10 - year policy - financial bonds. Insurance companies' buying power decreased significantly, securities firms' net buying of 5 - 10 - year Treasury bonds decreased sharply and sold a large amount of Treasury bonds over 10 years, and funds sold a large amount of Treasury bonds over 10 years and 5 - 10 - year policy - financial bonds that they had increased their positions before the festival [58][66]. - The make - up positions of small and medium - sized banks for 7 - 10 - year Treasury bonds were weak, securities firms had certain make - up operations and were the main buyers of 7 - 10 - year Treasury bonds, and funds generally sold 7 - 10 - year Treasury bonds last week. The make - up costs of main trading players were significantly different [69]. - Considering capital occupation and tax costs, commercial banks and insurance companies can obtain relatively higher returns by investing in local bonds [76]. 3.5 High - Frequency Data Tracking - Last week, the settlement price of rebar futures increased by 5.97% compared to before the festival, the settlement price of wire rod futures decreased by 5.71%, the settlement price of cathode copper futures increased by 2.04%, the cement price index decreased by 0.37%, and the Nanhua glass index increased by 2.02%. The CCFI index decreased by 4.00%, and the BDI index increased by 4.75% [83]. - In terms of food prices, the wholesale price of pork decreased by 2.53%, and the wholesale price of vegetables decreased by 5.02%. The settlement prices of Brent crude oil futures and WTI crude oil futures decreased by 1.41% and 1.78% respectively. The central parity rate of the US dollar against the RMB was 6.92 [83]. 3.6 Market Outlook - The bond market may show a volatile trend in the short term. Factors such as the potential increase in government bond supply after the Two Sessions in March, the seasonal increase in bank loan scale at the end of the quarter, and the lack of synergy among trading institutions make it unlikely for the yield of 10 - year Treasury bonds to break through the previous low. However, the probability may increase due to the rapid deterioration of the geopolitical situation [2][87].

Investment Rating - The report maintains a positive outlook on the innovative drug sector, highlighting multiple favorable developments and collaborations within the industry [1]. Core Insights - The innovative drug sector has seen significant collaborations, with a total business development (BD) value exceeding $50 billion in the first two months of 2026, indicating a robust growth trajectory [16]. - The report emphasizes the upcoming Q2 conferences (AACR, ELCC, ASCO, EHA) as potential catalysts for stock price increases in innovative drug companies due to expected positive data releases [16]. - The report identifies a strong performance in the medical consumables sub-sector, which has shown a 4.0% increase, while the best-performing sectors year-to-date include hospitals, medical consumables, and offline pharmacies [21][27]. Summary by Sections 1. Investment Strategy - The pharmaceutical index rose by 0.50% this week, underperforming the CSI 300 index by 0.58 percentage points, ranking 26th in industry performance [13]. - Year-to-date, the pharmaceutical sector has increased by 2.96%, outperforming the CSI 300 index by 1.21 percentage points, ranking 24th [21]. - The current valuation level (PE-TTM) for the pharmaceutical industry is 31.06 times, with a premium of 66.99% over the entire A-share market [23]. 2. Industry and Company News - Frontier Biotech has entered a global licensing agreement with GSK for small nucleic acid drugs, with a total potential value of $963 million, including an upfront payment of $40 million [14]. - Bai Li Tianheng's dual-target antibody ADC, iza-bren, has successfully met primary endpoints in a Phase III trial for triple-negative breast cancer (TNBC), indicating significant progress in this high-risk area [15]. - The report notes that several innovative drug companies have issued profit forecasts, suggesting a sustained interest and potential growth in the sector [16]. 3. Steady Portfolio - Recommended stocks include Heng Rui Medicine (600276), Bai Jie Shen Zhou-U (688235), Yi Fan Medicine (002019), and others, indicating a diversified approach to investment in the pharmaceutical sector [16][17].

Asset Allocation Model - The report introduces a dual-driven asset allocation model based on the Black-Litterman (BL) framework, integrating strategic and tactical approaches to enhance decision-making and investment performance[1] - The strategic component utilizes asset timing and regression analysis to generate posterior distributions of asset returns, improving the foresight of allocation decisions[1] - The tactical component focuses on style rotation strategies in the A-share market, dynamically tracking market style shifts to optimize investment portfolios[1] Timing Strategies - The bond timing strategy is based on economic fundamentals (economic growth, real estate cycle) and market interest rates (repo rates, government bond yields) to time government bonds[2] - The commodity timing strategy for gold incorporates financial, monetary, and commodity attributes, analyzing real interest rates, inflation expectations, risk aversion, and supply-demand relationships[2] - The stock timing strategy for A-shares evaluates liquidity (money market rates, credit expansion), international influences (China-US interest rate differentials, exchange rates), and valuation metrics[2] Style Rotation Strategies - The size and value rotation strategy examines macroeconomic changes and the performance of large versus small-cap stocks, utilizing indicators like credit spreads and foreign capital inflows to construct timing metrics[3] - The growth versus value rotation strategy is driven by macro liquidity and micro technical factors, with liquidity being the core driver of growth-value style rotation[3] Performance Metrics - From December 31, 2013, to December 31, 2025, the BL model strategy achieved an annualized return of 12.10%, a Sharpe ratio of 2.38, and a maximum drawdown of 4.72%[7] - The latest asset allocation weights as of December 31, 2025, include 4.54% in money market funds, 58.64% in government bonds, 18.00% in gold, and 18.82% in A-shares, with A-shares equally allocated between small-cap and growth styles[8]

Investment Rating - The report maintains an "Outperform" rating for the engineering machinery sector [1]. Core Insights - January 2026 saw strong sales growth in excavators and loaders, driven by a combination of replacement cycles and external demand. The outlook for 2026 is positive, supported by proactive fiscal policies, stabilization in the European construction sector, and sustained high demand in emerging markets. The report emphasizes the importance of focusing on domestic demand renewal and overseas expansion strategies [5][11]. - The engineering machinery index rose approximately 0.15% in January 2026, underperforming the Shanghai Composite Index by 3.61 percentage points [11]. - Excavator sales in January 2026 reached 18,708 units, a year-on-year increase of 49.5%, with domestic sales at 8,723 units (up 61.4%) and exports at 9,985 units (up 40.5%) [16]. - Loader sales for the same month totaled 11,759 units, reflecting a 48.5% year-on-year increase, with domestic sales of 5,293 units (up 42.8%) and exports of 6,466 units (up 53.4%) [16]. Summary by Sections Market Review - The engineering machinery index in January 2026 increased by about 0.15%, lagging behind major indices such as the Shanghai Composite and CSI 300 [11]. - The performance of different segments showed mixed results, with the average price-to-earnings (PE) ratios for complete machine manufacturing and components being 29 and 39, respectively [11]. Industry Tracking - The report highlights significant growth in excavator and loader sales, with excavators showing a 49.5% increase and loaders a 48.5% increase in January 2026 [16][18]. - The report notes that electric excavators and loaders are gaining traction, with electric loader sales reaching 2,990 units and a penetration rate of 25.43% [16]. Macro Dynamics - The manufacturing PMI for January 2026 was reported at 49.3%, indicating a slight contraction in manufacturing activity. However, production levels remain above the critical point, suggesting ongoing expansion in manufacturing [44]. - Infrastructure investment is supported by the issuance of special bonds amounting to approximately 367.7 billion yuan, a year-on-year increase of 79.5% [5]. Key Targets - Recommended key players in the sector include leading manufacturers such as Zoomlion (000157), Sany Heavy Industry (600031), and XCMG (000425), as well as core component suppliers like Hengli Hydraulic (601100) and Aidi Precision (603638) [5][54].

Quantitative Models and Construction Methods 1. Model Name: GAN_GRU Model - **Model Construction Idea**: The GAN_GRU model combines Generative Adversarial Networks (GAN) for feature generation and Gated Recurrent Unit (GRU) for time-series feature encoding to create a stock selection factor[4][13][14] - **Model Construction Process**: - **GAN Component**: - The GAN consists of a generator (G) and a discriminator (D). The generator learns the real data distribution and generates realistic samples, while the discriminator distinguishes between real and generated data[23][24] - Generator loss function: $$L_{G} = -\mathbb{E}_{z\sim P_{z}(z)}[\log(D(G(z)))]$$ where \(z\) is random noise, \(G(z)\) is the generated data, and \(D(G(z))\) is the discriminator's output probability for generated data being real[24][25] - Discriminator loss function: $$L_{D} = -\mathbb{E}_{x\sim P_{data}(x)}[\log D(x)] - \mathbb{E}_{z\sim P_{z}(z)}[\log(1-D(G(z)))]$$ where \(x\) is real data, \(D(x)\) is the discriminator's output probability for real data, and \(D(G(z))\) is the discriminator's output probability for generated data[27][29] - GAN training alternates between updating the generator and discriminator parameters through backpropagation[30] - The generator uses an LSTM model to preserve the sequential nature of input features, while the discriminator employs a CNN model to process the 2D structure of the generated features[33][37] - **GRU Component**: - Two GRU layers (GRU(128, 128)) are used, followed by an MLP (256, 64, 64) to output predicted returns (\(pRet\)) as the stock selection factor[22] - Input features include 18 price-volume characteristics (e.g., closing price, turnover rate) sampled over the past 40 days to predict cumulative returns for the next 20 trading days[14][18] - Data preprocessing includes outlier removal, standardization, and cross-sectional normalization[18] - Training is conducted semi-annually with rolling updates, using Adam optimizer, a learning rate of \(1e-4\), and IC as the loss function[18] - **Model Evaluation**: The GAN_GRU model effectively integrates GAN's feature generation capabilities with GRU's time-series encoding, making it suitable for capturing complex price-volume patterns in stock selection[4][13] --- Model Backtesting Results GAN_GRU Model - **IC Mean**: 0.1096*** (2019.02–2026.02)[41] - **ICIR (Non-Annualized)**: 0.87[42] - **Turnover Rate**: 0.82X[42] - **Recent IC**: -0.0105*** (latest period)[41][42] - **1-Year IC Mean**: 0.0517***[41][42] - **Annualized Return**: 38.13%[42] - **Annualized Volatility**: 23.18%[42] - **IR**: 1.64[42] - **Maximum Drawdown**: 27.29%[42] - **Annualized Excess Return**: 22.32%[41][42]

- The GAN_GRU factor is based on the GAN_GRU model, which utilizes Generative Adversarial Networks (GAN) for processing volume-price time series features and then employs the GRU model for time series feature encoding to derive the stock selection factor[4][13] - The GAN_GRU model includes two GRU layers (GRU(128, 128)) followed by an MLP (256, 64, 64), with the final output being the predicted return (pRet) used as the stock selection factor[22] - The GAN model consists of a generator and a discriminator. The generator aims to generate data that appears real, while the discriminator aims to distinguish between real and generated data. The generator's loss function is $L_{G} = -\mathbb{E}_{z\sim P_{z}(z)}[\log(D(G(z)))]$ and the discriminator's loss function is $L_{D} = -\mathbb{E}_{x\sim P_{d a t a}(x)}[\log D(x)] - \mathbb{E}_{z\sim P_{z}(z)}[\log(1-D(G(z)))]$[23][24][27] - The GAN_GRU model's training process involves alternating between training the generator and the discriminator until convergence[29][30] - The GAN_GRU factor's performance from February 2019 to February 2026 shows an IC mean of 0.1096, an annualized excess return of 22.32%, and an ICIR (non-annualized) of 0.87[41][42] - The latest IC value as of February 25, 2026, is -0.0105, with a one-year IC mean of 0.0517[41][42] - The top five industries for the GAN_GRU factor in February 2026, based on IC, are Electric Utilities, Retail, Real Estate, Construction, and Basic Chemicals, with IC values of 0.1257, 0.1196, 0.1151, 0.1130, and 0.1063, respectively[44] - The top five industries for the GAN_GRU factor over the past year, based on IC mean, are Steel, Computers, Media, Retail, and Food & Beverage, with IC means of 0.1404, 0.1175, 0.1132, 0.1014, and 0.0989, respectively[44] - The top five industries for the GAN_GRU factor's long positions in February 2026, based on excess returns, are Oil & Petrochemicals, Communications, Electronics, Non-ferrous Metals, and Computers, with excess returns of 7.91%, 3.11%, 3.06%, 2.78%, and 2.78%, respectively[45] - The top five industries for the GAN_GRU factor's long positions over the past year, based on average monthly excess returns, are Real Estate, Retail, Automobiles, Construction, and Consumer Services, with excess returns of 3.83%, 2.04%, 1.93%, 1.50%, and 1.49%, respectively[46] **Performance Metrics of GAN_GRU Factor:** - IC: 0.1096[41][42] - ICIR (non-annualized): 0.87[41][42] - Turnover Rate: 0.82X[41][42] - Recent IC: -0.0105[41][42] - One-year IC: 0.0517[41][42] - Annualized Return: 38.13%[41][42] - Annualized Volatility: 23.18%[41][42] - Information Ratio (IR): 1.64[41][42] - Maximum Drawdown: 27.29%[41][42] - Annualized Excess Return: 22.32%[41][42]