Risk-aware signal overlay
The design investigates risk attribution, regime detection, and position-level signal overlays alongside existing execution workflows.
Index / AssetModel
Structure on top
of noisy data.
An active-development research project exploring how structured signals can support theses, risk framing, attribution, and auditable portfolio decisions.
SYSCONCEPT
MKTSAMPLE DATA
FEEDSILLUSTRATIVE
MODELS7 ACTIVE
INFER14ms
QUEUE0
LAT42ms
UTC14:32:08.117
RISK · ARMED
NAV
$184.2M
+$1.14M day
P&L · DAY
+0.62%
+62bps
MTD
+3.41%
α +1.6 vs SPX
YTD
+18.2%
α +12.1 vs SPX
SHARPE · 1Y
2.14
net of fees
SORTINO
3.06
downside-adj
MAX DD
−4.8%
since incept
Positions · 14:32:08 UTC ● 7 of 24
SIDESYMTHESISWGTP&L D5DCONV
LONGSPYearnings dispersion3.4%+0.41
LONGXLFcurve normalization2.1%+0.18
SHORTIWMbreadth deterioration1.6%+0.09
LONGXLEsupply tightness1.4%−0.06
LONGBTCflow + macro confluence0.9%+0.22
SHORTHYGspread risk0.7%+0.03
LONGGLDreal-rate inflection0.6%+0.14
Equity · 1Y vs SPX
strategy+18.2%
α
+12.1
β
0.42
VOL
6.1%
CALMAR
3.79
Illustrative signal stream · ●
FLOWQQQ ask-side liquidity skew+0.412s
SENTcross-asset risk-on tilt+0.6211s
RISKvol-of-vol drift > floor−0.1828s
MACRO2s/10s steepening confirmed+0.2742s
EARNdispersion regime · week 4+0.341m
EXECrebalance proposed · sleeve EQ—1m
Inference graph · sleeve-eq · v2.41 ●
7 in6 · ReLU5 · ReLU4 out
Models · all sleeves
SLEEVE-EQv2.41 · acc 0.74 · drift 0.4σ
SLEEVE-RATESv1.18 · acc 0.71 · drift 0.6σ
SLEEVE-CREDITv0.92 · acc 0.69 · drift 1.2σ
SLEEVE-CRYPTOv3.04 · acc 0.66 · drift 0.3σ
REGIMEv2.07 · acc 0.81 · drift 0.2σ
RISK-OVLv4.12 · armed · last train 2h
Factor exposure
Equity
+0.64
Rates
−0.24
Credit
−0.16
FX
+0.12
Vol
−0.36
Crypto
+0.08
Sleeve correlation · 10×10
−1.00+1.0
Drawdown · 1Y
CURRENT
−0.2%
MAX
−4.8%
RECOV
14d
UPI
2.41
SPY524.18+0.41%
QQQ443.07+0.81%
IWM214.32−0.31%
XLF42.18+0.44%
XLE88.91−0.18%
GLD216.04+0.27%
TLT94.62−0.51%
HYG79.33−0.09%
BTC71,408+1.84%
DXY104.18−0.12%
VIX14.62+0.41%
2Y4.18%−2bp
10Y4.42%+1bp
SPY524.18+0.41%
QQQ443.07+0.81%
IWM214.32−0.31%
XLF42.18+0.44%
XLE88.91−0.18%
GLD216.04+0.27%
TLT94.62−0.51%
HYG79.33−0.09%
BTC71,408+1.84%
DXY104.18−0.12%
VIX14.62+0.41%
2Y4.18%−2bp
10Y4.42%+1bp
Research audience
Workflows under study.
The current research scope is aimed at teams evaluating auditable signal, risk, and portfolio workflows. It is not presented as a generally available production integration.
Sleeve-level discipline
The research asks how signal-to-thesis-to-allocation decisions can retain a reviewable record of why a position exists and what would invalidate it.
Risk-framed allocation
The research scope includes cross-sleeve allocation, confidence framing, and policy-based downside controls across several market categories.
Signal-to-execution
The design explores drift monitoring, latency budgets, and explicit human-approved controls between research signals and execution systems.
Design direction
A proposed signal-to-decision flow.
01 · Signals
Diagest or your own
Structured, provenance-aware inputs from Diagest or other research sources.
02 · Engine
Inference + risk overlay
Research models for confidence, regime context, factor exposure framing, and drift monitoring.
03 · Allocation
Decisions you can audit
A proposed output that keeps thesis labels, confidence, and decision lineage reviewable.
FAQ
Frequently asked.
What is AssetModel?
AssetModel is an active-development ixprt research project exploring how structured signals can support position theses, exposure framing, risk attribution, and auditable portfolio workflows.
How does AssetModel relate to Diagest?
The research direction is designed to consume structured, provenance-aware inputs such as those produced by Diagest. This describes an intended relationship rather than a generally available integration.
What markets are in the research scope?
The research scope includes equities, rates, credit, foreign exchange, volatility, and crypto. Coverage and production availability are not represented as finalized.
How is risk approached?
The design work emphasizes position-level confidence, factor exposure, correlation, drawdown controls, and decision lineage. The interface on this page is illustrative.
Is AssetModel generally available?
No public self-service product or standard integration package is represented as generally available. Contact ixprt for the current development status.
From the blog
Notes on what we're building.
Buying Guide
What is a Quant Engine? A Buyer's Guide for Funds and Family Offices
A definition of the quant-engine category, the four functions it covers, and what funds and family offices should evaluate.
ComparisonRisk Attribution Models Compared: Barra vs Axioma vs Custom Approaches
How the major risk-attribution approaches compare — and which fits which kind of fund.
Industry ReportThe State of AI in Finance: 2026 Benchmark
What shipped, what stuck, and where the puck is going across data infrastructure, quant integration, and AI research.