AT1 Bond Valuation¶

This notebook is a small, opinionated workflow for relative value screening in the AT1 universe using polars-bloomberg and interactive Altair charts.

We pull a bond universe (IDs, tickers, ratings, duration, back-end spread, z-spread) and then visualize how z-spread behaves as a function of:

Back-end spread (BE) and
Duration (Dur)

Both views include a simple fit. Bonds above the trend line are “cheap” on that axis (wider than the model would suggest), and bonds below the line are “rich”.

The two scatter plots are linked: clicking a point in one chart highlights the same bond in the other view, making it easy to spot names that look cheap (or rich) in both dimensions.

Out[1]:

In [2]:

Copied!





# Gett bond data from BQL
from polars_bloomberg import BQuery

query = """
let(
    #dur       = duration(duration_type=MODIFIED).value;
    #rtg_bb    = bb_composite().value;
    #zspread   = spread(spread_type=Z).value;
    #be        = flt_spread().value;
)
get(ticker, #dur, #rtg_bb, #zspread, #be)
for(
    filter(
        bondsUniv(types='active', consolidateDuplicates=true),
        ticker() in ['SHBASS', 'SEB', 'SWEDA', 'DANBNK', 'NDAFH', 'DNBNO']
        and crncy() == 'USD'
        and basel_iii_designation() == 'Additional Tier 1'
    )
)
"""

with BQuery() as bq:
    df = bq.bql(query).combine()
df.head(3)
# Gett bond data from BQL
from polars_bloomberg import BQuery

query = """
let(
    #dur       = duration(duration_type=MODIFIED).value;
    #rtg_bb    = bb_composite().value;
    #zspread   = spread(spread_type=Z).value;
    #be        = flt_spread().value;
)
get(ticker, #dur, #rtg_bb, #zspread, #be)
for(
    filter(
        bondsUniv(types='active', consolidateDuplicates=true),
        ticker() in ['SHBASS', 'SEB', 'SWEDA', 'DANBNK', 'NDAFH', 'DNBNO']
        and crncy() == 'USD'
        and basel_iii_designation() == 'Additional Tier 1'
    )
)
"""

with BQuery() as bq:
    df = bq.bql(query).combine()
df.head(3)

Out[2]:

shape: (3, 6)

ID	ticker	#dur	#rtg_bb	#zspread	#be
str	str	f64	str	f64	f64
"YR472577 Corp"	"DANBNK"	3.509404	"BBB-"	231.091119	259.9
"BR069680 Corp"	"SWEDA"	2.937217	"BBB"	270.448417	286.4
"BP504077 Corp"	"DANBNK"	0.394512	"BBB-"	154.481971	338.7