One question, start to finish.
Ask what if; Henselian runs the model on every record and every path; you read what then.
- What if
What if rates move by the six supervisory shocks?
- Run
Rate risk on the balance sheet, loan by loan
Every loan and deposit valued under each prescribed shock and on stochastic curves, not a duration shortcut.
- What then
Economic value of equity in each shock: which way the balance sheet leans, and by how much.
Start from a model that already asks.
Each is a working model in the app's Model builder, written in Hensel and run on the GPU. They start from a synthetic book; your data replaces it, record by record.
6 models in the Model builder
Treasury1
Rate risk on the balance sheet, loan by loan
Economic value of equity under the six supervisory shocks and stochastic curves, every loan and deposit valued.
Credit risk1
Credit losses, borrower by borrower
Expected loss and 99.9% credit VaR from a multi-factor model, every borrower's default probability conditional on the year.
Counterparty risk2
Counterparty exposure and CVA, nested
A swap and a floor with one counterparty, revalued at every future quarter on their own inner paths: CVA, expected exposure and the 95% peak exposure.
CVA on every trade
CVA and expected exposure, every swap revalued every quarter on every path, netted and collateralised by netting set.
Model engineering1
Calibrating the prepayment model on every loan
Maximum likelihood on every loan's quarterly history, slice by slice in parallel, then a backtest.
Group risk1
Group economic capital, every exposure simulated
Economic capital for the group and each silo from a t-copula, and how much diversification saves.
Fast enough to ask again.
for 1.5 trillion member-paths: a 2.5-million-member UK buy-in, end to end on one graphics card
faster than hand-optimised C on all 32 threads of the same machine
engine, one language, one app, for insurance, banking and energy alike
These were measured on an insurance run, a UK buy-in. The banking models above run on the same engine and language; a benchmark of their own is next.
One invitation
Come and challenge us.
If you work in insurance, banking, energy, or anywhere uncertainty drives important decisions, bring us the simulation problem your organisation wishes it could solve faster. Let's find out what's possible.