VWAP curve plays an important role in TradFi as it is the basis of many trading algorithms, with VWAP being one of the most common benchmark for execution algorithms. Being able to anticipate volumes going through intraday allows traders to be in line with the market and also take advantage of liquitidy.
Due to TradFi's open/close mechanisms, a "Volume Smile" is using formed, meaning there will be more liquidity at the ends of the open and close auction, forming a smile shape.
As crypto trades 24/7, we try to explore how the volume profile behaves in crypto.
The package allows you to easily retrieve historical volumes from Binance and construct their volume profile. load_historical_data pulls historical candlesticks from the exchange, generate_curves construct curves daily and aggregate them into an average curve.
The below code shows how we construct the cumulative curve for BTC Spot from Binance
from load_curves import load_historical_data, generate_curves
hist_df = load_historical_data('BTC/USDT')
cum_curve_df = generate_curves(hist_df)Intuitively, without open and close auctions, it is unlikely that a volume smile will be formed in crypto markets. However, it is worth noting that the first bin (UTC 0) stands out. Initial guess for the behavior is that similar to TradFi markets, there is often a fixing for portfolios and people often choose 00:00 as the rebalance time, which leads to a spike in volume.
A similar pattern is also observed for Perps.
However, the volume profile seems to exhibit weak seasonality in 8 hour intervals. Which is likely to be related to the settlement mechanism of perpetual futures (funding rates are paid out every 8 hours).