etf-premium

Fetch the ETF's regularMarketPrice and navPrice from yfinance, then compute (Price - NAV) / NAV x 100. A positive result is a premium, negative is a discount. Compare the result against the bid-ask spread to filter out market microstructure noise.

Loop through the ticker list, pull price and NAV for each via yf.Ticker().info, compute the premium percentage, and sort the results in a pandas DataFrame from deepest discount to highest premium. ETFs missing NAV data are flagged rather than silently dropped.

US large-cap equity ETFs like SPY normally stay within ±0.05%, while high-yield bond ETFs like HYG can run ±0.10-0.50% and crypto ETFs like IBIT or BITO commonly show 0.5-3% premiums. During market stress, bond ETF discounts have historically widened to 3-5%.

No. yfinance only exposes the most recent end-of-day navPrice, so historical premium time series cannot be built from it alone. For history, use ETF issuer websites, Bloomberg, or SEC N-PORT filings.

The excess move can be decomposed into a NAV-driven component (holdings weighted returns) and an excess premium driven by dealer gamma hedging, blocked AP arbitrage, or sentiment surges. Negative net GEX, spiking ATM IV, and concentrated call open interest indicate a gamma squeeze.

GEX depends heavily on the assumed dealer positioning convention, and public open interest alone cannot reveal actual dealer inventory. The dealer-share-of-volume figure is an upper-bound heuristic, and the NAV proxy is stale when underlying markets are closed.