- The multi-asset allocation model is constructed based on macro state recognition, incorporating growth/inflation factors, liquidity, and gold factors to create a dynamic risk budget portfolio [4][33][34] - The growth factor includes PMI, industrial added value, retail sales, fixed asset investment, and export data, while the inflation factor uses CPI and PPI. Liquidity factor is measured by M1 year-on-year growth [34][35] - Equity market characteristics are monitored using ERP (Equity Risk Premium), EP (Earnings Yield), and BP (Book-to-Price ratio) to construct stock-bond cost-effectiveness factors [34][35] - Gold investment factors are constructed using the dollar index, central bank gold purchases, and exchange rates to assess dynamic allocation value [34][35] - The model employs a multi-objective optimization approach, integrating asset momentum into traditional risk parity and risk budget frameworks. ETFs are used for portfolio construction, with dynamic adjustments based on macro signals [37][38] - The industry rotation model incorporates six dimensions: macro, financial, analyst expectations, ETF share changes, public fund/selected fund position momentum, and event momentum [39][41] - The industry rotation model has achieved an annualized return of 28% since 2012, with an annualized excess return of 18.1% over industry equal weight and a monthly excess win rate of 70% [42][43] - The industry rotation ETF strategy employs a five-layer recursive solution method to enhance portfolio performance, achieving an annualized return improvement of over 12% [77][78] - The "Accompanying Style Enhanced FOF" uses a dynamic multi-factor model focusing on Alpha and crowding factors, with quarterly adjustments to optimize fund selection and portfolio construction [46][47] - The "Accompanying Broad-based Enhanced FOF" employs a relative benchmark strategy to control tracking error while maximizing composite factor scores, using a dynamic multi-factor model [53][54] - The "Long-term Capability Factor FOF" combines Brinson model-based decomposition with TM and H-M models for timing and selection capabilities, incorporating style factors for enhanced fund selection [64][66] - The "KF-Alpha+ Trading FOF" uses quarterly data and Kalman filter-based industry estimation to construct Alpha factors, focusing on industry-specific stock selection capabilities [70][73] - The industry rotation ETF portfolio achieved a monthly excess return of 1.5% during the reporting period, with a full-period annualized excess return of 17.79% and an IR of 1.72 [78][79][87]