Memecoin Shiba Inu saw its price grow by more than 300 percent within one and the same month, marking a trading fury reminiscent of Dogecoin in early 2021. Indeed, the SHIB coin ranked as one of the biggest crypto in the world based on 24h trading volume in October 2021 - with trading activity being almost as high as that of Ethereum. Originally starting out as a fun token based on the famous Japanese dog breed, the digital currency grabbed mainstream attention in 2021 probably due to its low barrier of entry: Even after its price spikes, Shiba Inu was one of the cheapest coins available, especially when compared to the price of 99 other cryptocurrencies. Combine this with survey information that most consumers in the U.S. and the UK invest in crypto either for growth or fun, and Shiba Inu is probably seen by many as a relatively simple, "quick win" investment. As of May 19, 2025, the price of one million Shiba Inu is 14.08.

Somewhat resembling developments of Dogecoin in early 2021, the market cap of Shiba Inu (SHIB) more than tripled in only a few weeks. This rapid growth especially occurred since September 2021, with the cryptocurrency breaking the ** billion U.S. dollar barrier multiple in the following month. Whilst no exact reason was given why the memecoin grew so fast during this time, the digital currency did become available on Coinbase, one of the world's most well-known crypto exchanges, on September 17. This move, and the fact that consumers in the U.S. and the UK overall tend to invest in crypto for fun or as a potential growth investment, might explain the popularity of the dog-themed coin. Indeed, Shiba Inu ranked as the number ** cryptocurrency in the world based on market cap in late October 2021 - closely behind DeFi-affilated like Uniswap (UNI) and Terra (LUNA).

Shiba Inu trading volume accelerated significantly in October 2021, as the *** trade volume grew from roughly *********** U.S. dollars into double digits. Whilst the cryptocurrency based on the famous Japanese dog breed already had a trading peak early in ********, ************ could be seen as the breakthrough month for the coin as it became one of the most valuable cryptocurrencies available. The meme-inspired digital currency saw a major price increase during this month, as it invoked a similar sentiment as to early 2021 when Dogecoin became a popular cryptocurrency to invest in.

All-time high price data for Shiba Inu, including the peak value, date achieved, and current comparison metrics.

Shiba Inu reached its all-time high price on October 28, 2021.

Find my notebook : Advanced EDA & Data Wrangling - Crypto Market Data where I cover the full EDA and advanced data wrangling to get beautiful dataset ready for analysis.

Find my Deep Reinforcement Learning v1 notebook: "https://www.kaggle.com/code/franoisgeorgesjulien/deep-reinforcement-learning-for-trading">Deep Reinforcement Learning for Trading

Find my Quant Analysis notebook:"https://www.kaggle.com/code/franoisgeorgesjulien/quant-analysis-visualization-btc-v1">💎 Quant Analysis & Visualization | BTC V1

Dataset Presentation:

This dataset provides a comprehensive collection of hourly price data for 34 major cryptocurrencies, covering a time span from January 2017 to the present day. The dataset includes Open, High, Low, Close, Volume (OHLCV), and the number of trades for each cryptocurrency for each hour (row).

Making it a valuable resource for cryptocurrency market analysis, research, and trading strategies. Whether you are interested in historical trends or real-time market dynamics, this dataset offers insights into the price movements of a diverse range of cryptocurrencies.

This is a pure gold mine, for all kind of analysis and predictive models. The granularity of the dataset offers a wide range of possibilities. Have Fun!

Ready to Use - Cleaned and arranged dataset less than 0.015% of missing data hour: crypto_data.csv

First Draft - Before External Sources Merge (to cover missing data points): crypto_force.csv

Original dataset merged from all individual token datasets: cryptotoken_full.csv

crypto_data.csv & cryptotoken_full.csv highly challenging wrangling situations: - fix 'Date' formats and inconsistencies - find missing hours and isolate them for each token - import external data source containing targeted missing hours and merge dataframes to fill missing rows

see notebook 'Advanced EDA & Data Wrangling - Crypto Market Data' to follow along and have a look at the EDA, wrangling and cleaning process.

Date Range: From 2017-08-17 04:00:00 to 2023-10-19 23:00:00

Date Format: YYYY-MM-DD HH-MM-SS (raw data to be converted to datetime)

Data Source: Binance API (some missing rows filled using Kraken & Poloniex market data)

Crypto Token in the dataset (also available as independent dataset): - 1INCH - AAVE - ADA (Cardano) - ALGO (Algorand) - ATOM (Cosmos) - AVAX (Avalanche) - BAL (Balancer) - BCH (Bitcoin Cash) - BNB (Binance Coin) - BTC (Bitcoin) - COMP (Compound) - CRV (Curve DAO Token) - DENT - DOGE (Dogecoin) - DOT (Polkadot) - DYDX - ETC (Ethereum Classic) - ETH (Ethereum) - FIL (Filecoin) - HBAR (Hedera Hashgraph) - ICP (Internet Computer) - LINK (Chainlink) - LTC (Litecoin) - MATIC (Polygon) - MKR (Maker) - RVN (Ravencoin) - SHIB (Shiba Inu) - SOL (Solana) - SUSHI (SushiSwap) - TRX (Tron) - UNI (Uniswap) - VET (VeChain) - XLM (Stellar) - XMR (Monero)

Date column presents some inconsistencies that need to be cleaned before formatting to datetime: - For column 'Symbol' and 'ETCUSDT' = '23-07-27': it is missing all hours (no data, no hourly rows for this day). I fixed it by using the only one row available for that day and duplicated the values for each hour. Can be fixed using this code:

start_timestamp = pd.Timestamp('2023-07-27 00:00:00')
end_timestamp = pd.Timestamp('2023-07-27 23:00:00')

hourly_timestamps = pd.date_range(start=start_timestamp, end=end_timestamp, freq='H')

hourly_data = {
  'Date': hourly_timestamps,
  'Symbol': 'ETCUSDT',
  'Open': 18.29,
  'High': 18.3,
  'Low': 18.17,
  'Close': 18.22,
  'Volume USDT': 127468,
  'tradecount': 623,
  'Token': 'ETC'
}

hourly_df = pd.DataFrame(hourly_data)
df = pd.concat([df, hourly_df], ignore_index=True)

df = df.drop(550341)

• Some rows for 'Date' have extra digits '.000' '.874' etc.. instead of the right format YYYY-MM-DD HH-MM-SS. To clean it you can use the following code:

# Count the occurrences of the pattern '.xxx' in the 'Date' column
count_occurrences_before = df['Date'].str.count(r'\.\d{3}')
print("Occurrences before cleaning:", count_occurrences_before.sum()) 

# Remove '.xxx' pattern from the 'Date' column
df['Date'] = df['Date'].str.replace(r'\.\d{3}', '', regex=True)

# Count the occurrences of the pattern '.xxx' in the 'Date' column after cleaning
count_occurrences_after = df['Date'].str.count(r'\.\d{3}')
print("Occurrences after cleaning:", count_occurrences_after.sum()) 

**Disclaimer: Any individual or entity choosing to engage in market analysis, develop predictive models, or utilize data for trading purposes must do so at their own discretion and risk. It is important to understand that trading involves potential financial loss, and decisions made in the financial mar...

Dogecoin reached its all-time high price on May 7, 2021.