量化配置研究系列四：日股量化择时模型构建：由日本股汇负相关引发的日股定价探讨

Quantitative Models and Construction Methods Model Name: OFR Financial Stress Index - Construction Idea: The index aims to measure financial stress by incorporating various financial indicators, including exchange rates and interest rates of developed economies[1] - Construction Process: The index includes the 10-Year US Treasury yield, 10-Year German yield, GOLD/USD Real Spot Exchange Rate, USD/JPY Spot Exchange Rate, USD/Swiss Franc Spot Exchange Rate, and the US Dollar Index[2] - Evaluation: The index has certain timing effectiveness in financial stress scenarios[1] Model Name: Comprehensive Timing Signal - Construction Idea: The model uses signals from six dimensions: Japanese economy, overseas economy and trade, exchange rates, employment, consumption, and inflation[6] - Construction Process: - Japanese Economy: Combined signals from Japan Manufacturing PMI and Japan Services PMI - Overseas Economy and Trade: Combined signals from US Manufacturing PMI, US Services PMI, and trade balance - Exchange Rate Expectation: Difference between US-Japan interest rates and USD/JPY exchange rate - Employment: Combined signals from employee compensation and labor force population - Consumption: Combined signals from average savings rate of households with two or more workers and monthly consumption expenditure - Inflation: CPI-PPI difference[7] - Evaluation: The model shows a comprehensive timing effectiveness with a total win rate of 65.85%, a long win rate of 64.29%, a short win rate of 72.00%, a payoff ratio of 1.11, and an excess return of 4.60%[6][7] Model Backtesting Results 1. Comprehensive Timing Signal: - Annualized Return: 14.65% - Nikkei 225: 10.05% - Excess Return: 4.60% - Drawdown: 24.60% - Volatility: 17.06% - Sharpe Ratio: 0.89 - Calmar Ratio: 0.60[11] Quantitative Factors and Construction Methods Factor Name: Japanese Economy - Construction Idea: Reflects domestic economic growth through PMI indicators[6] - Construction Process: Combined signals from Japan Manufacturing PMI and Japan Services PMI - Formula: $\text{Economic Signal} = \text{Manufacturing PMI} + \text{Services PMI}$ - Evaluation: The factor has a total win rate of 59.06%, a long win rate of 62.30%, a short win rate of 51.02%, a payoff ratio of 1.08, and an excess annualized return of 0.16%[7] Factor Name: Overseas Economy and Trade - Construction Idea: Reflects the impact of US economic activities on Japanese trade[6] - Construction Process: Combined signals from US Manufacturing PMI, US Services PMI, and trade balance - Formula: $\text{Overseas Signal} = \text{US Manufacturing PMI} + \text{US Services PMI} + \text{Trade Balance}$ - Evaluation: The factor has a total win rate of 61.00%, a long win rate of 65.71%, a short win rate of 54.72%, a payoff ratio of 0.95, and an excess return of -1.39%[7] Factor Name: Exchange Rate Expectation - Construction Idea: Reflects the impact of interest rate differentials and exchange rate movements on Japanese stocks[6] - Construction Process: Difference between US-Japan interest rates and USD/JPY exchange rate - Formula: $\text{Exchange Rate Signal} = \text{US-Japan Interest Rate Differential} - \text{USD/JPY Exchange Rate}$ - Evaluation: The factor has a total win rate of 56.98%, a long win rate of 58.78%, a short win rate of 45.83%, a payoff ratio of 1.20, and an excess return of 0.92%[7] Factor Backtesting Results 1. Japanese Economy: - Total Win Rate: 59.06% - Long Win Rate: 62.30% - Short Win Rate: 51.02% - Payoff Ratio: 1.08 - Excess Annualized Return: 0.16%[7] 2. Overseas Economy and Trade: - Total Win Rate: 61.00% - Long Win Rate: 65.71% - Short Win Rate: 54.72% - Payoff Ratio: 0.95 - Excess Return: -1.39%[7] 3. Exchange Rate Expectation: - Total Win Rate: 56.98% - Long Win Rate: 58.78% - Short Win Rate: 45.83% - Payoff Ratio: 1.20 - Excess Return: 0.92%[7]