Overall this is a wonderful idea given how critical the property market is for so many people, affecting their livelihoods and their wellbeing. Would be valuable to get a deeper understanding of the trends etc.
Keep/Cut:
The introduction discussion of the foreign buyers and the 15% tax - not sure if that's really necessary. It's an interesting topic in and of itself, but it is not the most salient to your later outline of the issues. I think you can get into your topic without necessarily mentioning this or at least really tighten up the story.
The idea of 'what's hot, what's not' is a very interesting concept and I think very catchy / worth hanging on to.
Re-arrange/Add:
Need to clarify the "objective function" or "dependent variable" ie what problem or decision are you gathering all this intelligence to solve?
Add in the beginning of the middle some kind of mini-intro e.g. "The three key things that drive (x) are demographics/lifestyle; city planning investment and local enterprise initiatives; along with location and geography. It comes up later, and you arrive at it inductively, later in the section, but might be good to state upfront.
The question of "period validity" needs to be tightened up. I think you will be able to capture the time based issue if you are calculating mortgate and interest rates at several time periods (t) to creat the model.
The time period might be an interesting "parameter" to provide the end users eg., select when you will be active in the real-estate market - 1 year, 2 years, 3 years etc.
This project would lend itself very well to a strong map-based (geo-coded) data visualization application.
Overall this is a wonderful idea given how critical the property market is for so many people, affecting their livelihoods and their wellbeing. Would be valuable to get a deeper understanding of the trends etc.
Keep/Cut:
Re-arrange/Add: