Compare Databento vs. CME end-of-day settlement volumes

compare_cme_eod.py

from enum import Enum

import databento as db
import pandas as pd


class Stat(Enum):
    SETTLEMENT_VOLUME = 6

client = db.Historical()

data = client.timeseries.get_range(
    dataset="GLBX.MDP3",
    schema="statistics",
    stype_in="parent",
    symbols=["SR3.OPT"],
    start="2024-10-16",     # CME publishes previous trade date's settlement volume on next day
    end="2024-10-17",
)

df = data.to_df()

df = df[df["symbol"].str.startswith("SR3Z4")]
df = df[df.stat_type == Stat.SETTLEMENT_VOLUME.value]
df = df.groupby("symbol").last()[["quantity"]].rename({"quantity": "volume_databento"}, axis="columns")

def symbol_rename(s):

    strike = s.split(" ")[0]
    put_call = "C" if s.endswith("Call") else "P"
    return f"SR3Z4 {put_call}{strike}" 

# Official statistics from:
# https://www.cmegroup.com/markets/interest-rates/stirs/three-month-sofr.volume.options.html
# https://www.cmegroup.com/CmeWS/exp/voiProductDetailsViewExport.ctl?media=xls&tradeDate=20241015&reportType=F&productId=8462
# Expiration:   Dec 2024
# Trade date:   Tuesday, 15 Oct 24
# Last updated: 16 Oct 2024 09:02:47 AM CT
df_cme = pd.read_csv("cme_voi_details.csv", sep="\t", thousands=",")
df_cme = df_cme[["strike", "volume_total"]]
df_cme = df_cme.rename({"strike": "symbol", "volume_total": "volume_cme"}, axis="columns")
df_cme["symbol"] = df_cme["symbol"].apply(symbol_rename)
df_cme["volume_cme"] = df_cme["volume_cme"].astype("Int64")

df_merged = pd.merge(df, df_cme, on="symbol", how="left")
df_merged.to_csv("comparison-databento-vs-cme.csv", index=False)

print(df_merged)