Explore Basis Trade Data

import pandas as pd
import polars as pl

from matplotlib import pyplot as plt
import seaborn as sns
sns.set()

from settings import config
import load_bases_data

DATA_DIR = config("DATA_DIR")
MANUAL_DATA_DIR = config("MANUAL_DATA_DIR")

Arbitrage Spreads

df = load_bases_data.load_combined_spreads_wide(data_dir=MANUAL_DATA_DIR)
# df = df.ffill(limit=5).dropna()
# df.info()

df.info()

<class 'pandas.core.frame.DataFrame'>
DatetimeIndex: 5563 entries, 2001-12-04 to 2024-02-28
Data columns (total 32 columns):
 #   Column             Non-Null Count  Dtype  
---  ------             --------------  -----  
 0   Box_06m            4435 non-null   float32
 1   Box_12m            4423 non-null   float32
 2   Box_18m            4435 non-null   float32
 3   CDS_Bond_HY        4280 non-null   float32
 4   CDS_Bond_IG        4437 non-null   float32
 5   CIP_AUD            5563 non-null   float32
 6   CIP_CAD            5456 non-null   float32
 7   CIP_CHF            5562 non-null   float32
 8   CIP_EUR            5562 non-null   float32
 9   CIP_GBP            5561 non-null   float32
 10  CIP_JPY            5493 non-null   float32
 11  CIP_NZD            5374 non-null   float32
 12  CIP_SEK            4894 non-null   float32
 13  Eq_SF_Dow          5106 non-null   float32
 14  Eq_SF_NDAQ         5090 non-null   float32
 15  Eq_SF_SPX          5088 non-null   float32
 16  TIPS_Treasury_02Y  4687 non-null   float32
 17  TIPS_Treasury_05Y  4687 non-null   float32
 18  TIPS_Treasury_10Y  4687 non-null   float32
 19  TIPS_Treasury_20Y  4687 non-null   float32
 20  Treasury_SF_02Y    4211 non-null   float32
 21  Treasury_SF_05Y    4575 non-null   float32
 22  Treasury_SF_10Y    4801 non-null   float32
 23  Treasury_SF_20Y    4836 non-null   float32
 24  Treasury_SF_30Y    2412 non-null   float32
 25  Treasury_Swap_01Y  5563 non-null   float32
 26  Treasury_Swap_02Y  5551 non-null   float32
 27  Treasury_Swap_03Y  5515 non-null   float32
 28  Treasury_Swap_05Y  5515 non-null   float32
 29  Treasury_Swap_10Y  3900 non-null   float32
 30  Treasury_Swap_20Y  3106 non-null   float32
 31  Treasury_Swap_30Y  3106 non-null   float32
dtypes: float32(32)
memory usage: 738.8 KB

df.tail()

	Box_06m	Box_12m	Box_18m	CDS_Bond_HY	CDS_Bond_IG	CIP_AUD	CIP_CAD	CIP_CHF	CIP_EUR	CIP_GBP	...	Treasury_SF_10Y	Treasury_SF_20Y	Treasury_SF_30Y	Treasury_Swap_01Y	Treasury_Swap_02Y	Treasury_Swap_03Y	Treasury_Swap_05Y	Treasury_Swap_10Y	Treasury_Swap_20Y	Treasury_Swap_30Y
date
2024-02-22	NaN	NaN	NaN	NaN	NaN	8.251404	17.293579	17.858582	5.813416	8.302063	...	-16.325012	-74.609985	-211.763977	4.56	-13.77	-21.139999	-29.330000	-41.220001	-69.800003	-78.449997
2024-02-23	NaN	NaN	NaN	NaN	NaN	6.583923	18.220825	12.748291	4.431885	8.169006	...	-15.239990	-59.401978	-190.694977	6.46	-13.30	-20.480000	-29.780001	-42.310001	-71.440002	-77.029999
2024-02-26	NaN	NaN	NaN	NaN	NaN	5.156189	18.380493	10.448364	2.971252	8.405518	...	-14.094971	-65.319977	-183.165985	5.58	-12.50	-20.879999	-28.650000	-42.099998	-71.699997	-77.980003
2024-02-27	NaN	NaN	NaN	NaN	NaN	6.320984	17.697937	18.856445	6.171143	8.237305	...	-18.984985	-63.867981	-194.381989	5.60	-12.95	-22.200001	-30.320000	-43.040001	-72.839996	-79.169998
2024-02-28	NaN	NaN	NaN	NaN	NaN	8.978394	18.759705	27.148132	8.660278	8.211975	...	-12.888000	-79.199005	-191.681000	4.94	-12.23	-21.049999	-28.900000	-42.540001	-71.650002	-77.550003

5 rows × 32 columns

df.reindex(sorted(df.columns), axis=1).filter(regex=("Treasury_SF_*")).info()

<class 'pandas.core.frame.DataFrame'>
DatetimeIndex: 5563 entries, 2001-12-04 to 2024-02-28
Data columns (total 5 columns):
 #   Column           Non-Null Count  Dtype  
---  ------           --------------  -----  
 0   Treasury_SF_02Y  4211 non-null   float32
 1   Treasury_SF_05Y  4575 non-null   float32
 2   Treasury_SF_10Y  4801 non-null   float32
 3   Treasury_SF_20Y  4836 non-null   float32
 4   Treasury_SF_30Y  2412 non-null   float32
dtypes: float32(5)
memory usage: 152.1 KB

df.reindex(sorted(df.columns), axis=1).filter(regex=("Treasury_SF_*")).plot()

<Axes: xlabel='date'>

(
    df
    .loc["2023":,:]
    .reindex(sorted(df.columns), axis=1)
    .filter(regex=("Treasury_SF_*"))
    .plot()
)
plt.ylim([-300,200])

(-300.0, 200.0)

(
    df
    .loc["2024-07":,:]
    .reindex(sorted(df.columns), axis=1)
    .filter(regex=("Treasury_SF_*"))
    .plot()
)
plt.ylim([-300,200])

(-300.0, 200.0)

df.reindex(sorted(df.columns), axis=1).filter(regex=("Treasury_SF_*")).plot()
plt.ylim([-300,200])

(-300.0, 200.0)

WINDOW = 30
ax= df.reindex(sorted(df.columns), axis=1).filter(regex=("TIPS_*")).rolling(WINDOW).mean().plot()
# plt.ylim([-100,100])
box = ax.get_position()
ax.set_position([box.x0, box.y0, box.width * 0.8, box.height])

# Put a legend to the right of the current axis
ax.legend(loc='center left', bbox_to_anchor=(1, 0.5))

plt.ylabel(f"Spread (bps), {WINDOW}-day Rolling Mean")
plt.title("TIPS-Treasury Arbitrage Spread");

WINDOW = 30
ax= df.reindex(sorted(df.columns), axis=1).filter(regex=("Treasury_SF_*")).rolling(WINDOW).mean().plot()
# plt.ylim([-100,100])
box = ax.get_position()
ax.set_position([box.x0, box.y0, box.width * 0.8, box.height])

# Put a legend to the right of the current axis
ax.legend(loc='center left', bbox_to_anchor=(1, 0.5))

plt.ylabel(f"Spread (bps), {WINDOW}-day Rolling Mean")
plt.title("Treasury Cash Futures Arbitrage Spread\nFutures-implied risk-free rate minus OIS rate");

WINDOW = 30
ax= df.reindex(sorted(df.columns), axis=1).filter(regex=("Treasury_Swap_*")).rolling(WINDOW).mean().plot()
# plt.ylim([-100,100])
box = ax.get_position()
ax.set_position([box.x0, box.y0, box.width * 0.8, box.height])

# Put a legend to the right of the current axis
ax.legend(loc='center left', bbox_to_anchor=(1, 0.5))

plt.ylabel(f"Spread (bps), {WINDOW}-day Rolling Mean")
plt.title("Treasury Swap Arbitrage Spread");

WINDOW = 30
ax= df.reindex(sorted(df.columns), axis=1).filter(regex=("CIP_*")).rolling(WINDOW).mean().plot()
# plt.ylim([-100,100])
box = ax.get_position()
ax.set_position([box.x0, box.y0, box.width * 0.8, box.height])

# Put a legend to the right of the current axis
ax.legend(loc='center left', bbox_to_anchor=(1, 0.5))

plt.ylabel(f"Spread (bps), {WINDOW}-day Rolling Mean")
plt.title("CIP Arbitrage Spread");

df

	Box_06m	Box_12m	Box_18m	CDS_Bond_HY	CDS_Bond_IG	CIP_AUD	CIP_CAD	CIP_CHF	CIP_EUR	CIP_GBP	...	Treasury_SF_10Y	Treasury_SF_20Y	Treasury_SF_30Y	Treasury_Swap_01Y	Treasury_Swap_02Y	Treasury_Swap_03Y	Treasury_Swap_05Y	Treasury_Swap_10Y	Treasury_Swap_20Y	Treasury_Swap_30Y
date
2001-12-04	NaN	NaN	NaN	NaN	NaN	52.145401	NaN	13.314026	9.184860	11.789230	...	NaN	NaN	NaN	7.60	NaN	NaN	NaN	NaN	NaN	NaN
2001-12-05	NaN	NaN	NaN	NaN	NaN	46.005219	NaN	12.012970	5.848602	15.825348	...	NaN	NaN	NaN	8.10	NaN	NaN	NaN	NaN	NaN	NaN
2001-12-06	NaN	NaN	NaN	NaN	NaN	49.061096	NaN	16.064270	10.815582	12.939957	...	NaN	NaN	NaN	4.80	NaN	NaN	NaN	NaN	NaN	NaN
2001-12-07	NaN	NaN	NaN	NaN	NaN	58.456711	NaN	15.164368	10.896408	21.666672	...	NaN	NaN	NaN	5.30	NaN	NaN	NaN	NaN	NaN	NaN
2001-12-10	NaN	NaN	NaN	NaN	NaN	53.464584	NaN	12.022736	8.434219	14.334442	...	NaN	NaN	NaN	2.20	NaN	NaN	NaN	NaN	NaN	NaN
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
2024-02-22	NaN	NaN	NaN	NaN	NaN	8.251404	17.293579	17.858582	5.813416	8.302063	...	-16.325012	-74.609985	-211.763977	4.56	-13.77	-21.139999	-29.330000	-41.220001	-69.800003	-78.449997
2024-02-23	NaN	NaN	NaN	NaN	NaN	6.583923	18.220825	12.748291	4.431885	8.169006	...	-15.239990	-59.401978	-190.694977	6.46	-13.30	-20.480000	-29.780001	-42.310001	-71.440002	-77.029999
2024-02-26	NaN	NaN	NaN	NaN	NaN	5.156189	18.380493	10.448364	2.971252	8.405518	...	-14.094971	-65.319977	-183.165985	5.58	-12.50	-20.879999	-28.650000	-42.099998	-71.699997	-77.980003
2024-02-27	NaN	NaN	NaN	NaN	NaN	6.320984	17.697937	18.856445	6.171143	8.237305	...	-18.984985	-63.867981	-194.381989	5.60	-12.95	-22.200001	-30.320000	-43.040001	-72.839996	-79.169998
2024-02-28	NaN	NaN	NaN	NaN	NaN	8.978394	18.759705	27.148132	8.660278	8.211975	...	-12.888000	-79.199005	-191.681000	4.94	-12.23	-21.049999	-28.900000	-42.540001	-71.650002	-77.550003

5563 rows × 32 columns

# ax= (
#     df
#     .reindex(sorted(df.columns), axis=1)
#     .filter(regex=("CIP_*"))
#     .loc["2024-07":,:]
#     .dropna()
#     .plot()
# )

# plt.ylim([-100,100])
# box = ax.get_position()
# ax.set_position([box.x0, box.y0, box.width * 0.8, box.height])

# # Put a legend to the right of the current axis
# ax.legend(loc='center left', bbox_to_anchor=(1, 0.5))

# plt.ylabel(f"Spread (bps)")
# plt.title("CIP Arbitrage Spread");