Description

With commercial property, AVMs stand to benefit a wide range of important, but labor-intensive processes; preliminary valuations, underwriting, portfolio valuations, collateral assessment, risk management and more. However, a property valuator that combines traditional valuation methods with his subjective opinion leads to biased and time-consuming valuations, thus making it inappropriate for large portfolios.

Infodim LP, is creating an Automated Valuation Model (AVM) that offers objective, fast, and accurate property valuation using state-of-the-art models, such as Statistical and Machine Learning models.

Our model is using two methods, the conventional comparative approach and a machine-based approach.

Firstly, the comparative approach has been modeled to perform data-driven valuations in large portfolios using cutting-edge statistical models, and according to global valuation standards. This method is using all the needed data points combined with spatial analysis for each property.

Secondly, the machine-based method uses hundreds of data points from multiple sources to perform accurate and unbiased valuations in large scale.

The Machine Learning model preprocesses, enriches and normalizes all the available data for each property and combines it with macro-scale data points that results in a very fast and accurate valuation, feature extraction, market analysis and risk management.