Predictive analysis & asset management
Vaste Rendébrance continuously aggregates market data and applies artificial intelligence models to produce numerical recommendations. We make analysis methods previously reserved for institutional managers available to middle-income families.
Start my analysis See public resultsProof by facts
We do not publish performance promises. We publish a history: the date of each recommendation, the suggested allocation, and its actual evolution on the markets, good or bad.
| Recommendation | Horizon | Status |
|---|---|---|
| Bond/equity rebalancing | 18 months | Currently being monitored |
| Reduction of cyclical sector exposure | 6 months | Closed |
| International diversification | 36 months | Currently being monitored |
Illustrative extract from newspaper format; the real data and their complete history are accessible from the platform.
Our reliability is based on two complementary disciplines. The backtesting consists of comparing each model to past market data, to verify its behavior before putting it into production. The stochastic modeling then introduces random market scenarios, in order to estimate a range of plausible outcomes rather than a single, misleading forecast. These two methods do not guarantee any future results; on the other hand, they reduce the risk of decisions based on intuition alone.
What the analysis actually changes
The following three mechanisms work together: they monitor, alert, then adjust the allocation strategy in line with family objectives.
The models ingest continuous market feeds and recalculate portfolio scenarios as soon as a significant variation is detected. The goal is not speed for its own sake, but the ability to spot a signal before it translates into a visible loss on a bank statement.
Each household defines a risk tolerance threshold, expressed in acceptable volatility and not in abstract jargon. The system monitors the gap between this threshold and the portfolio's actual exposure, and flags any drift before it affects precautionary savings or medium-term plans.
The proposed asset allocation takes into account the horizon of each objective – retirement, studies, transfer – rather than applying a single model. The risk-weighted return is recalculated each time there is a change in maturity or family situation.
The analysis engine, step by step
Three separate steps, each verifiable, provide insight into why a recommendation is made rather than just an end result.
Market prices, macroeconomic indicators and regulatory publications are collected and standardized in a common format, so that the model works on comparable and precisely dated data.
Several independent models separately assess likely market scenarios. Their results are compared, and excessive divergences between models trigger a review before any recommendation.
The final recommendation is accompanied by the hypotheses retained and the scenarios discarded, so that the household understands the reasoning and not just the conclusion.
Project yourself into use
These examples illustrate the type of signal that the analysis can bring up, without anticipating a guaranteed result.
Retirement preparation
When a capital withdrawal is planned for a fixed deadline, the system flags unstable market periods likely to coincide with this deadline, to consider a gradual adjustment of the allocation rather than a last minute decision.
Funding of studies
A college goal has a specific deadline. The analysis adjusts the share of volatile assets as maturity approaches, to limit exposure at times when deferral flexibility is lowest.
Capital preservation
A portfolio can appear diversified on the surface while remaining focused on the same underlying risk factor. The model highlights these hidden correlations between seemingly distinct investment lines.
Our approach
Vaste Rendébrance was built around a simple principle: a household that understands why a recommendation is made makes a more confident decision than a household that simply carries it out. Each analysis module documents its assumptions and limitations.
We work with public market data and industry standard quantitative models, which we adapt for family rather than institutional use: risk thresholds expressed in everyday language, horizons based on real projects, and performance logs that can be consulted without prior accounting.
Access to the platform gives rise to an initial portfolio diagnosis, based on the objectives and the horizon that you provide.