El Salvador EMBI directed graphs

Monthly — is the current spread priced?

Current EMBI	272 bp (2026-09-02)
Macro fair value	330 bp (month-end 2026-08-31)
Rich / cheap	very_rich
Gap (current − fair)	-58 bp (-1.595σ)
Month-end residual	-59 bp
Residual σ	36 bp
In-sample R²	0.955

4 monthly print(s) may not yet have been tradable at the last EMBI date.

Parent	Δ last month	Contrib bp	Typical print lag (d)
tot_gap (t)	-0.000	+10.5	45
us_hy_oas (t)	-0.220	-7.5	0
dxy (t)	-0.956	+5.6	0
tot (t)	0.000	+0.0	45
debt_gap (t)	0.000	+0.0	35
current_account (t)	0.000	+0.0	25

Has EMBI had time to trade the prints?

Input	Est. release	Traded after print?	Days since
us_hy_oas	2026-08-31	yes	2
tot	2026-10-15	pending	-43
tot_gap	2026-10-15	pending	-43
dxy	2026-08-31	yes	2
debt_gap	2026-10-05	pending	-33
current_account	2026-09-25	pending	-23

Monthly DAG

Narrative edges (strongest first)

From		To	Lag	β (z)	Child R²
public_debt_gdp	→	debt_gap	t	+1.000	0.116
nominal_deficit	→	primary_balance	t	+0.816	-0.100
copper	→	tot_gap	t	-0.739	-144.224
copper	→	tot	t	-0.581	-1296.742
nominal_deficit	→	public_debt_gdp	t	-0.543	-9.673
debt_gap	→	embi_bra	t	+0.526	-0.974
public_debt_gdp	→	embi_bra	t	+0.526	-0.974
dxy	→	embi_bra	t	+0.518	-0.974
current_account	→	embi_bra	t	-0.506	-0.974
tot	→	current_account	t	+0.448	-0.114
tot_gap	→	embi_bra	t	-0.392	-0.974
tot	→	reserves	t	+0.382	-7.491
copper	→	exports_growth	t	+0.366	-2.666
tot_gap	→	current_account	t	+0.356	-0.114
primary_balance	→	public_debt_gdp	t	-0.355	-9.673
primary_balance	→	debt_gap	t	-0.355	0.116
reserves	→	reserves_growth	t	+0.336	0.049
copper	→	embi_bra	t	+0.313	-0.974
current_account	→	reserves_growth	t	+0.260	0.049
tot	→	exports_growth	t	-0.214	-2.666
us_hy_oas	→	embi_bra	t	-0.207	-0.974
current_account	→	reserves	t	+0.176	-7.491
exports_growth	→	current_account	t	+0.110	-0.114
tot	→	primary_balance	t	-0.055	-0.100
reserves_growth	→	embi_bra	t	-0.054	-0.974
primary_balance	→	embi_bra	t	-0.052	-0.974
tot_gap	→	primary_balance	t	-0.044	-0.100
tot	→	embi_bra	t	-0.030	-0.974

ΔEMBI engine (holdout score)

Ridge on ΔEMBI: 1 risk (HY or VIX or S&P) + 1 rates/USD + 1 commodity + 1 local + 1 stock. Ranked by holdout R², not quintile lift. Best: theory|lags=[0, 1, 2] · parents: us_hy_oas, dxy, copper, exports_growth, public_debt_gdp · 182 train / 19 test through 2026-09-30.

From		To	Lag	β (z)	Child R²
dxy	→	embi_bra	t	+0.170	0.355
us_hy_oas	→	embi_bra	t	+0.143	0.355
exports_growth	→	embi_bra	t	+0.111	0.355
copper	→	embi_bra	t−2	+0.089	0.355
us_hy_oas	→	embi_bra	t−1	+0.084	0.355
exports_growth	→	embi_bra	t−1	+0.079	0.355
copper	→	embi_bra	t−1	+0.059	0.355
exports_growth	→	embi_bra	t−2	-0.058	0.355
dxy	→	embi_bra	t−2	-0.047	0.355
us_hy_oas	→	embi_bra	t−2	+0.040	0.355
copper	→	embi_bra	t	-0.033	0.355
dxy	→	embi_bra	t−1	-0.000	0.355
public_debt_gdp	→	embi_bra	t	-0.000	0.355
public_debt_gdp	→	embi_bra	t−1	-0.000	0.355
public_debt_gdp	→	embi_bra	t−2	-0.000	0.355

Method	Lags	OOS R²	σ-quintile	Lift vs naive	Naive middle	±1 bin	3-class	Sign	RMSE Δ	Edges	Parents
theory|lags=[0, 1, 2] ← best	[0, 1, 2]	0.285	47.4%	-0.053	52.6%	94.7%	47.4%	73.7%	0.230	10	15
pc|lags=[0, 1, 2]	[0, 1, 2]	0.285	47.4%	-0.053	52.6%	94.7%	47.4%	73.7%	0.230	10	15
pc_strict|lags=[0, 1, 2]	[0, 1, 2]	0.285	47.4%	-0.053	52.6%	94.7%	47.4%	73.7%	0.230	10	15
tan|lags=[0, 1, 2]	[0, 1, 2]	0.285	47.4%	-0.053	52.6%	94.7%	47.4%	73.7%	0.230	9	15
hybrid|lags=[0, 1, 2]	[0, 1, 2]	0.285	47.4%	-0.053	52.6%	94.7%	47.4%	73.7%	0.230	16	15
theory|lags=[0, 1]	[0, 1]	0.177	57.9%	0.053	52.6%	100.0%	57.9%	68.4%	0.247	10	10
pc|lags=[0, 1]	[0, 1]	0.177	57.9%	0.053	52.6%	100.0%	57.9%	68.4%	0.247	10	10
pc_strict|lags=[0, 1]	[0, 1]	0.177	57.9%	0.053	52.6%	100.0%	57.9%	68.4%	0.247	10	10
tan|lags=[0, 1]	[0, 1]	0.177	57.9%	0.053	52.6%	100.0%	57.9%	68.4%	0.247	9	10
hybrid|lags=[0, 1]	[0, 1]	0.177	57.9%	0.053	52.6%	100.0%	57.9%	68.4%	0.247	16	10
theory|lags=[1]	[1]	-0.066	36.8%	-0.158	52.6%	89.5%	36.8%	73.7%	0.281	10	5
pc|lags=[1]	[1]	-0.066	36.8%	-0.158	52.6%	89.5%	36.8%	73.7%	0.281	10	5
pc_strict|lags=[1]	[1]	-0.066	36.8%	-0.158	52.6%	89.5%	36.8%	73.7%	0.281	10	5
tan|lags=[1]	[1]	-0.066	36.8%	-0.158	52.6%	89.5%	36.8%	73.7%	0.281	9	5