Qimtav — AI-powered financial data analysis dashboard

Analyzing financial data with artificial intelligence to support documented investment decisions

Qimtav analyzes real-time market data and economic indicators, and links each recommendation to a public scorecard that can be reviewed before making a decision.

Public performance record Real-time data update Verifiable predictive models Documented methodology
Qimtav — a working interface for analyzing financial portfolio data
Work methodology

A methodology based on data, not expectations

Qimtav relies on predictive models trained on historical and real-time data, and produces recommendations with a clear confidence score rather than absolute numbers. The goal is to reduce reliance on guesswork in long-term financial decisions.

Each recommendation goes through a subsequent validation phase, and its outcome is recorded in a public scorecard available for review, allowing the methodology to be evaluated rather than relying on generic promises.

Technical basis

How does the system reduce the degree of uncertainty in a financial decision?

Three processing layers work sequentially: predictive modeling, real-time processing, and matching to the user's personal goals.

Predictive modeling

Predictive modeling based on historical and real-time data

The system processes time series of prices and economic indicators to generate probability ranges for future outcomes, rather than a single fixed number. Each recommendation is accompanied by a confidence score that shows how stable the model is at that moment.

Data sourceMarket + economic indicators
Output typeProbability range
Trust indexAttached to each recommendation
Real-time processing

Real-time processing of market changes

Inputs are constantly updated with each market impact change, minimizing the time lag between a change occurring and its reflection in a recommendation. This limits exposure to decisions based on delayed data.

Update cycleContinuous
Processing scopeActive market indicators
GoalReduce time lag
Strategic matching

Match recommendations with the user's financial goals

Modeling results are linked to user data: time horizon, risk tolerance, and investment objective. The final recommendation reflects these data rather than a uniform general result for all users.

User inputHorizon + risk tolerance
processingMatching with model results
He came outPersonalized recommendation
Results achieved

A documented performance record that can be reviewed

The table below shows the published log format. The complete record includes all recommendations issued and the date each result was verified.

The record was last updated: It is reviewed monthly and the verification date is pinned at the top of each row. Download the full record (PDF)
Period Recommendation type The result achieved Verification status
First quarter 2024 Asset allocation Within the expected range Verified
Second quarter 2024 Portfolio rebalancing Higher than the expected minimum Verified
Third quarter 2024 Volatility alert Identical to the high risk scenario Verified
Fourth quarter 2024 Sectoral diversification Under final review Under review

The values above are an illustrative example of the published data format. The complete history includes every recommendation issued by the system with verification details.

Mechanism of action

Three steps from raw data to decision

A consistent sequence keeps it clear where each recommendation comes from and why.

01

Data collection

The system collects market data, economic indicators and the user's account data from pre-defined sources.

02

Analysis and synthesis

Predictive models combine this data to generate potential scenarios, and accompany each scenario with an estimated risk ratio.

03

Decision outcomes

The user gets a prioritized recommendation, with a brief explanation of the factors that influenced the score.

Risk management

Dedicated units to reduce exposure to long-term risks

Four independent units work together to protect the financial decision from unconsidered fluctuations.

Volatility alerts

Real-time monitoring of unusual market movements, and immediate notification when pre-set user limits are exceeded.

Portfolio optimization

A proposed reallocation of assets based on the target risk ratio and the specified investment time horizon.

Market sentiment analysis

Measuring the general trend from multiple text data sources to estimate buying or selling pressure on a specific asset.

Diversification drive

Identifying excessive concentration in a single sector or asset, and proposing alternatives to reduce correlation between portfolio components.

Transparency

Frequently asked questions about methodology and data

How is my data stored?

Account data is stored on encrypted servers and is not shared with third parties for marketing purposes. The user can request deletion of his data at any time by contacting the support team.

How is the accuracy of recommendations calculated?

Each recommendation is compared to the actual outcome after a pre-determined period of time, and is then classified as “achieved” or “under review” in the public record described in the Scorecard section.

How does the subscription system work?

The subscription level determines the scope of access to the different analysis modules. Details of levels and prices are available upon direct contact with the sales team.

Start by reviewing your performance history before making your next financial decision

You can review the format of published data first, or start a live analysis of your financial situation.