Time to start sharing (automatically) data I collcect. Following an article https://github.com/ctberthiaume/gdcp I installed gdcp on my Raspberry Pi and finalized configuration with allowing access to my Google Drive to this app.
It took 10 hours to calculate model = ARIMA(history, order=(20,1,0)). Results are merged onto 1 screen, where transparent green line represents results from above model. Red line comes from previous calculations from model = ARIMA(history, order=(7,1,0))
So, moving to my laptop worked. I got a bit faster environment than old good Raspberry Pi. Unfortunately (or, rather fortunately) I discovered a bug in my previous codes – this beautiful picture “predicting” next 300 transactions.. Was bullshit 🙂 After carefully reading code I realized it predicts/forecasts just 1 next value and retrains the model.
As I’m running more and more often out of memory when playing with ARIMA on Raspberry I decided to install Python and all libraries under Windows. Unfortunately I run into some problems with statsmodels package and trying to figure this out.
Wow. I’m excited!
My first attemp to truly predict price changes on bitcoin. Let’s see how it goes.
Yesterday’s example made me think a lot. Below is equation for a correlation coefficient between two data sets:
Today I wanted to prove/disprove if there’s a correlation between palladium daily price changes and accordingly bitcoin price changes. Unfortunately correlation between these two data sets does not exist. Excel Correl function was at the level of 0.03. It very much suprises me as correlation between daily price of both palladium and bitcoin is at the level of 0.88.
I’m trying to understand why is that..
Thanks for reading,