Featured Photo by Rūdolfs Klintsons
Let’s look at the Dogecoin:
Dogecoin (DOGE) is traded on CRYPTO Exchanges. Dogecoin is peer-to-peer digital currency powered by the Blockchain technology. Dogecoin is one of many evolving digital currencies in which encryption is used to regulate the generation of units of currency and verify the transactions independently of a central authority. It is traded on 36 exchanges in multiple currencies.
Dogecoin is trading at 0.088 as of the 14th of November 2022, a 3.65 percent up since the beginning of the trading day. Dogecoin has more than 62 % chance of experiencing financial distress in the next few years of operation. It also did not have a very good performance during the last 90 trading days.
Based on a normal probability distribution, the odds of Dogecoin to move above the current price in 90 days from now is about 26.56% (This Dogecoin probability density function shows the probability of Dogecoin Crypto Coin to fall within a particular range of prices over 90 days). Source: Macroaxis.
Dogecoin has been active in the last 3 months, and it is presently trading with a bearish sentiment. Cryptocurrencies such as Dogecoin are digital assets that allow for secure payments and are represented by ledger entries internal to the system, generally referred to as a blockchain. Blockchain implementations use encryption algorithms and cryptographic techniques that safeguard entries in the ledger. Cryptocurrency assets such as Dogecoin are becoming very popular among investors and have been praised for their portability, inflation resistance, and transparency.
Let’s look at the ML Dogecoin price prediction using the public-domain historical data from 2014-09-17 to 2021-09-03.
Let’s import the libraries and read the input data stored in the directory YOURPATH
!pip install autots
from autots import AutoTS
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
from seaborn import regression
data = pd.read_csv(“YOURPATH/Dogecoin.csv”)
print(“Shape of Dataset is: “,data.shape,”\n”)
Shape of Dataset is: (2544, 7) Date Open High Low Close Adj Close Volume 0 2014-09-17 0.000293 0.000299 0.000260 0.000268 0.000268 1463600.0 1 2014-09-18 0.000268 0.000325 0.000267 0.000298 0.000298 2215910.0 2 2014-09-19 0.000298 0.000307 0.000275 0.000277 0.000277 883563.0 3 2014-09-20 0.000276 0.000310 0.000267 0.000292 0.000292 993004.0 4 2014-09-21 0.000293 0.000299 0.000284 0.000288 0.000288 539140.0
Let’s drop NaNs and plot the data
plt.title(“DogeCoin Price INR”)
Let’s train the ML model and perform the 10-day AutoTS forecast
model = AutoTS(forecast_length=10, frequency=’infer’, ensemble=’simple’, drop_data_older_than_periods=200)
model = model.fit(data, date_col=’Date’, value_col=’Close’, id_col=None)
prediction = model.predict()
forecast = prediction.forecast
print(“DogeCoin Price Prediction”)
DatepartRegression with avg smape 10.57
This ML backtesting algorithm can help crypto traders optimize and improve their strategies, find any technical or theoretical flaws, as well as gain confidence in their strategy before applying it to the real world markets.