机构：中金公司 研究员：严佳卉/吕松涛/张帅帅 1H25 营业收入符合我们预期，归母净利润好于我们预期东亚银行公布1H25 业绩。公司1H25 营业收入 同比下降2.1%，符合我们预期，主要由于HIBOR 下行压缩息差空间；公司1H25 归母普通股净利润同比 增长24.7%，表现好于我们预期，主要由于减值损失少提。 利率下行拖累净利息收入。1H25 公司净利息收入同比下降10.7%，环比2H24 下行11.5%，下滑幅度略 高于已披露财报同业。主要由于公司信贷敞口集中于中国香港与中国大陆，两地降息及在内地执行相对 保守投放策略压制了息差。向前看，HIBOR 在8 月13 日后快速上行，截至8 月21 日1MHIBOR 已恢复 至2.84%，我们认为净利息收入压力下半年或有所缓解。 发展趋势 1H25 公司中国大陆境内客户同比增长62%，南下客户同比增长54%，带动零售银行业务手续费收入同 比增长285%。 风险 中国大陆及中国香港地产相关敞口资产质量超预期下行。 信用成本自高位有所回落，是净利润超预期的主要原因，后续展望保持审慎。公司近3 年来首次出 现"信贷损失少提+不良贷款率下行+拨备覆盖率微降"组合，1 ...

Group 1 - HSBC Research has lowered the target price for Bank of Communications (03328.HK) from HKD 6.6 to HKD 6.3, maintaining a "Hold" rating [1] - The bank reported a year-on-year increase of 1.5% in earnings per share for the first quarter, attributed to better net interest margin trends and reduced credit costs compared to peers [1] - However, the growth in costs was higher than that of competitors, and the upcoming capital injection is expected to dilute earnings per share and dividends [1] Group 2 - HSBC has adjusted its earnings per share forecasts for the company downwards by 1.2%, 3.3%, and 2.7% for the years 2023 to 2027 respectively [1]

公司1Q25 业绩符合我们预期 公司1Q25 净利润/拨备前利润/营业收入同比+1.4%/-4.5%/-1.0%，业绩符合我们预期。 发展趋势 利润增速上升。公司1Q25 净利润/拨备前利润/营业收入同比增速较2024 年分别+0.4ppt/-4.6ppt/-1.9ppt， 净利润增速改善，主要由于信用成本下降，资产减值损失同比下降13.5%；营业收入增速有所下降，其 中其他非息收入同比下降10.6%，较2024 年下降15.8ppt，主要由于债券及权益市场波动加大导致；利息 净收入同比增长2.5%，较2024 年下降1.0ppt，主要由于一季度息差重定价压力较大。 资产增长较快。公司一季度总资产/信贷同比增长7.4%/8.7%，较2024 年分别上升1.4ppt/1.2ppt，资产保 持稳定扩张。重点领域看，公司科技金融信贷较年初增幅+11.3%，节能降碳企业信贷增幅+7.5%，普惠 小微贷款增幅+5.9%，养老产业信贷增幅+13.8%，均保持较快增长。公司预计2025 年全年人民币信贷 增幅与2024 年持平，结构上继续投向五篇大文章等重点领域，并继续提升零售信贷占比。 维持盈利预测与估值不变。当前A 股 ...

Quantitative Models and Construction Methods - **Model Name**: Huatai Three-Cycle Model **Model Construction Idea**: The model analyzes the price movement of COMEX gold settlement prices using three classic economic cycles: Kitchin, Juglar, and Kuznets cycles. It identifies the dominant cycle components influencing gold price trends[17] **Model Construction Process**: 1. The model decomposes the year-on-year sequence of COMEX gold settlement prices into three cycle components: Kitchin, Juglar, and Kuznets cycles 2. The amplitude of the extracted cycle components is ranked as Kuznets > Juglar > Kitchin 3. The current positions of the Kuznets and Juglar cycles are analyzed to predict future gold price trends[17][19] **Model Evaluation**: The model highlights that gold prices are more influenced by longer-term cycles (Kuznets and Juglar) compared to shorter-term cycles (Kitchin), providing insights into the strong cyclical positioning of gold in the current market[17] Model Backtesting Results - **Huatai Three-Cycle Model**: The model indicates that the Kuznets cycle is near its peak, and the Juglar cycle is in an upward phase, suggesting that gold prices are likely to remain strong in the near term[17][19] Quantitative Factors and Construction Methods - **Factor Name**: Gold as a Portfolio Stabilizer **Factor Construction Idea**: Gold is evaluated as a low-correlation asset with high long-term returns, making it a potential stabilizer in diversified investment portfolios[3][21] **Factor Construction Process**: 1. Historical performance of gold is compared with other major asset classes (e.g., equities, bonds, commodities) over different time horizons (1 year, 5 years, 10 years, 20 years) 2. Risk-return metrics such as Sharpe ratio, Calmar ratio, and maximum drawdown are calculated for gold and other assets 3. Correlation analysis is conducted to assess gold's relationship with other asset classes[21][23][24] **Factor Evaluation**: Gold demonstrates high returns, low volatility, and low correlation with other assets, making it a valuable addition to investment portfolios for risk diversification and return enhancement[21][23] - **Factor Name**: Gold in Asset Allocation Portfolios **Factor Construction Idea**: The impact of adding gold to a traditional stock-bond portfolio is analyzed to evaluate its contribution to portfolio performance[3][24] **Factor Construction Process**: 1. A baseline portfolio is constructed with 60% bonds (ChinaBond New Comprehensive Wealth Index) and 40% stocks (CSI A500 Index) 2. Two new portfolios are created by reallocating 10% of the baseline portfolio to gold (AU9999 spot gold): - Portfolio A: 50% bonds, 40% stocks, 10% gold - Portfolio B: 60% bonds, 30% stocks, 10% gold 3. Monthly rebalancing is applied, and backtesting is conducted over the period from January 3, 2005, to February 19, 2025 4. Risk-return metrics (e.g., annualized return, Sharpe ratio, maximum drawdown) are calculated for all portfolios[24][26] **Factor Evaluation**: Adding gold improves portfolio Sharpe ratios and reduces volatility, demonstrating its role as a stabilizing asset in diversified portfolios[26] Factor Backtesting Results - **Gold as a Portfolio Stabilizer**: - Sharpe Ratio: 0.59 (AU9999 spot gold), 0.57 (London spot gold), 0.56 (COMEX gold futures) - Maximum Drawdown: -44.88% (AU9999 spot gold), -44.62% (London spot gold), -44.52% (COMEX gold futures) - Annualized Return: 9.07% (AU9999 spot gold), 9.76% (London spot gold), 9.74% (COMEX gold futures)[23] - **Gold in Asset Allocation Portfolios**: - Portfolio A: Annualized Return 7.17%, Sharpe Ratio 0.72, Maximum Drawdown -35.47% - Portfolio B: Annualized Return 6.69%, Sharpe Ratio 0.88, Maximum Drawdown -26.86% - Baseline Portfolio: Annualized Return 6.63%, Sharpe Ratio 0.68, Maximum Drawdown -33.36%[26]