Message from @Von Bolt
Discord ID: 626051738339639307
The way you can make cycle theory works is by considering the fact that there are these periods when innovation can give you higher NPVs for the same investment that others.
It is actually quite simple-there are these breakthroughs which create new industries, like the computer for example.
And if you put money in this sector you can expect faster returns than in any other sector.
This is why capital tend to flow into them.
This helps them grow and they can provide even greater returns with the new capital they accumulate until they reach a technological barrier of some kind where the return diminishes.
At which point you need new tech to start things back up
There is just no way the same profit margins can be extracted out of the sector any more and the people start loosing money instead of gaining them.
However there is already tension build in the economy because people don't know the highest point of the cycle has been reached yet.
It is very difficult to determine it actually.
And this is where credits default and investment becomes insecure.
If you could determine it it would just cause it to happen sooner, unless you didn't act on it
s' poor form to interfere with your observational data
However, if the country's or the global economy relies heavily on growth in these sectors, cause they are the most innovative, too, so it makes sense putting your eggs in this basket, this fall of the profit margin can take with it a greater chunk of the economy.
And there is why crises happen.
Quite elegant, actually, however, it is a pain in the ass to find enough data to prove it and to actually build a model around the idea.
I think the problem is that too much data has been collected to prove it which was acted upon, same as with a lot of the big data stuffs that companies love so much
You have to run an observational study because an experimental one would be atrociously complicated, unless you were to model it
<:ahegao:462286952335671296>
@Hexidecimark Oh, I do want to build a model-problem is I do not know what data to use and where to find it cause such things as the GDPs or patent numbers are too general for me.
You don't have to use existing data, you have to make realistic actor units and a reality-reflective medium for them to interact in
What I need are the profit margins per sector for the last 100 years or so and a fine method to put into considerations the relationship between the investment for patent in them chronologically.
And this is where the pain begins cause there are hoards of data that aren't available out there.
Start with one industry
Most companies don't publish sensitive R&D information.
This is what is actually preventing me from gaining some falsifiability.
And I can't just get my hands on the accounting of big companies, if you get what I mean?
@Ardith Prime Metokur Youtuber Career in a nutshell
That's why you just start with one
Still am very frustrated, thou.
I just can't get enough detailed data to start something, or may be can't sort out what I already have.
But there is this thing, that I'm already busy enough doing the many other things I already have then to get a data science degree, too.
Don't take it all head on at once