Data always says something unless it’s randomly generated. At the very least Chinese data provides lower bounds on some things. You can get somewhat better estimates if you model their incentives (though the lying will greatly increase the uncertainty and complexity of any model)
Data always says something unless it’s randomly generated. At the very least Chinese data provides lower bounds on some things. You can get somewhat better estimates if you model their incentives (though the lying will greatly increase the uncertainty and complexity of any model)
-