I am not a data scientist or any kind of expert, this is ONLY my opinion based on running a few experiments.
Special thank you to whoever sent me the data https://mega.nz/file/RQ4gjZxK#0jOb0ZyyjWMVGk9sRUvHhjlMa5X–dJvP3S2CurE_XQ it looks like they found every price and merchant number and compiled it by date, what a champion! Anyone else that feels like donating their hard work 👉🏻 [email protected] 👈🏻
When you start a data science project you come up with a hypothesis and then try and prove (or disprove) said hypothesis.
For example when I first looked at https://bitcash.io/review/qoin-price-compilation/ it looked like the price was growing fast early then slowing. After trying to fit a curve to the real data, it became obvious that was false. What I failed to notice was the Y axis (on the left, up and down) in the early screenshots was $0.20 top to bottom whereas the later more recent charts were $1.00 – $1.50. A typical data mistake.
The first thing I wanted to know is “can you fit a curve to the price data” suggesting the price is artificial and after quite a few attempts discovered https://arachnoid.com/polysolve/ where you can copy data directly from Excel and do a Polynomial Regression Data Fit. This was the result first 2 degrees, then 3 degrees (x axis Day #, y axis Price):
I am used to dealing in statistical significance and had to google “correlation coefficient polynomial regression” but the results really surprised even me.
The merchant numbers are similar but while there are 180 price data points there are only 9 merchant numbers so I will not share any opinions. This can be explored later when we talk about how many Qoin are “minted” on average per merchant.
Python is the data scientists language and again I hope someone out there on the world wide web can take the data and reveal something obvious after analysis. In the meantime I will keep on chugging on!
One more thing I wanted to hypothesize, the genesis block (blockchain language for the first block) doesn’t seem to correlate (for price purposes) to January. It appears to be more like early April or March which is SUPER critical to my “1% interest on average a day compounded it’s obviously a scam” theory. Food for thought 🙂
👉🏻 [email protected] 👈🏻 just sayin’