CASE STUDIES / RISK · ANALYSIS · AUG 2025
Monte Carlo raises the same VaR 40%.
Cargo-level value at risk on a representative LNG book, computed twice. History and simulation disagree by forty percent, which makes the method itself a risk decision.
$46M → $64M
5% PORTFOLIO VAR · HISTORICAL VS MONTE CARLO
Four legs, four indices.
01 · THE BOOK
Two FOB longs and two DES shorts, each on a different index, run over twelve cargoes on two 171k vessels. Forward profit before any risk overlay: $169.2M.
12 CARGOES · 3.5-3.9M MMBTU EACH · FORWARD PROFIT $169.2M
Two methods, one book.
02 · THE METHODS
Historical risk uses realised prices from January 2024 to July 2025; Monte Carlo runs one thousand simulated price sets. Simulation is uniformly higher across VaR and expected shortfall at both confidence levels, because it reaches tails history never produced.
Historical
Monte Carlo
5% VAR
1% VAR
5% ES
1% ES
Simulation reaches beyond history.
03 · THE GAP
Across all four metrics the simulated view adds roughly forty to fifty percent. A desk that reports only historical VaR understates exactly the scenarios that hurt most.
+39.7%
5% VAR · MC VS HISTORICAL
+41.1%
1% VAR
+48.2%
1% EXPECTED SHORTFALL
+48.4%
5% EXPECTED SHORTFALL
Re-indexation trims the tail.
04 · THE FIX
Shifting the UK legs from 0.8x NBP to 0.5x NBP plus $4 holds profit flat within a fifth of a percent and cuts portfolio VaR by nearly a tenth. One cargo gets riskier in exchange: risk moves, it rarely disappears.
05 · THE VERDICT
Choosing the method is itself a risk decision: the same book reports $46M or $64M depending on the lens.
CONSTRUCTED EXAMPLE BOOK · INDEX VALUES OF 20 JULY 2025 · NO RE-OPTIMISATION ACROSS PATHS
THIS ANALYSIS RAN ON X-LNG
The whole analysis, priced inside the optimisation that plans your fleet.
X-LNG runs analyses like this on whole portfolios in seconds, so the answer sits in every netback, diversion and charter call rather than in a spreadsheet weeks later.
