The process of creating forecasts at ALFA is carried out in four stages, designed to ensure the forecast is as accurate as possible and maximizes the value and utility that can be derived from it in the future.

Stage 1: Defining the Forecast
At this stage, the specific variable we aim to predict is defined, along with the time frame for the forecast. The goal is to create a clear focus for the forecasting process.

Stage 2: Building an Impact Network
In this stage, an initial layer of data is constructed, containing factors that directly influence the central variable we aim to predict. For each factor in the primary layer, an additional layer of data is created to represent the factors that influence it. This process is repeated for every factor at every layer, creating a multi-layered data structure. The deeper and more detailed the structure, the more accurate the forecasts become.

Stage 3: Exposure to Information
At this stage, the model is exposed to available and up-to-date information, which updates the data layers. As a result, the forecast for the central variable is updated and becomes more accurate.

Stage 4: Applying the Forecast
At this stage, we define the exact action to take in each scenario based on the forecast.

By utilizing forecasts, ALFA can allocate resources and actions in the present to maximize the value and benefits achieved in the future.For example, using forecasts can increase future returns on capital:  
‍
Suppose the market valuation of a public company is $100 million. The forecast predicts that within five years, one of two scenarios will occur:  
1. An 80% probability that the company's value will grow to $500 million.  
2. A 20% probability that the company's value will drop to $50 million.
ALFA can purchase shares of the company in the stock market with an expected return of 310% over five years.  
Additionally, ALFA can reduce the risk of loss from 20% to 4%, while maintaining the same expected return, by splitting its capital between two separate forecasts.

Currently:  
1. Model V1 generates forecasts for the VX futures contract.  
2. Model S1 generates forecasts for the ES futures contract.  

In the future:  
1. More accurate models for V1 and S1.  
2. Models forecasting assets not traded within the S&P 500 index.  
3. Models forecasting assets that are not publicly traded at all.

Currently:
Investors can connect their brokerage accounts to ALFA's technologies (API), subject to the legal terms and regulations in each country.  

In the future:  
A publicly traded financial instrument will be launched, enabling investors to follow ALFA's model forecasts by investing in it.

Currently:
Managing non-tradable assets based on ALFA's forecasts is not possible.

In the future:
ALFA will develop advanced models designed to predict the performance of non-tradable assets, enabling effective management of such assets.

ALFA is looking for passionate researchers to develop advanced models for more accurate and efficient forecasting. If you see yourself as a good fit for such a role, we’d be delighted to hear from you.

- Aharon Ben Hamo – CEO and Director  
- Raz Biton – Forecast Manager and Director  
- Raz Leshem – Forecast Manager and Director  
- Yoel Bretstein – Investor and Director  
- Dr. Amir Elaluf – Director

The purpose of the website is to provide organizations and investors with relevant and up-to-date information about ALFA and the models.

No, the capital is held with brokers who comply with SEC, FINRA, and SIPC regulations. ALFA serves as a third party in the transaction and connects to the broker via API.