These are publicy on-going research and codes to my master's thesis entitled "A Machine Learning Approach to Predicting China's Industrial Policy."
Many economies, governments, and businesses are closely watching China’s every economic policy to plan appropriate responses to political relationships and economic agreement with China. Government policy in China is set by ‘steering committees’, in both party and state at all levels of government. Policy is formed by the party, set into administrative regulation by the state bureaucracy and finally molded into legislation for passage through the National People's Congress. Current literature have used machine learning to predict policy change with China’s news media People’s Daily, but none have used online ministry data to directly predict China’s industrial policy movement. The policy tightening and expansion are defined by frequency of keywords associated the market-reform vocabulary. Unlike the previous mentioned literature, models in this paper are more focused on short-term industrial policy tightening or expanding movements by yearly quarters, as oppose to long term structural changes. This paper presents the first China Industrial Policy Index (CIPI), which tracks the key industrial sectors closely associated with China’s industrial policy agendas; each will be given individual indexes to track its movement. The CIPI is composed of industrial sectors that the Government is focused on tightening. Theoretically, this is an applied-quantitative research to explain the behavior of ministries, gain policy intelligence insights, and predict China’s Industrial Policy tightening and control.
Research Question: With China increasingly publishing industrial policy programs, how can stakeholders track and predict China’s Industrial Policy?
How? Through scraped ministry online data to train Natural Language Process Model to analyze the relationship between ministry online data and other metadata signals to predict tightening or loosening of a China’s Industrial Policy?
- How measure the tightening and loosening of policy?
- How to track the relationship between media and state?
- Not about predicting a certain policy coming out
- Not about 100% accurate in predicting the time window of the policy publishing.