The error is a statistically derived estimate of the reliability of the price estimate, which is defined using the distribution of transactions. A star rating is derived from the error, and the more stars it gets, the more accurate the rating is
Each price and market rent estimate provided by the model is given a star rating and an estimated percentage of error. The error is defined as follows = The difference between the indicated market value (the price estimate given by the service) and the actual market value (for example, the debt-free price of a trade that took place on the free market).
In practice, there is never a corresponding completed transaction of the object being evaluated according to the value date, so the real error is unknown. However, we can derive the expected value of the error from other stores. When the estimated error is, for example, 10%, with about 68% probability the estimate is ±10%.
A reliability rating is derived from the estimated error, which is described with stars. The more stars a review has, the more reliable it is. The classes are as follows:
Five stars = 0 - 10%
Four stars = 10 - 15%
Three stars = 10 - 20%
Two stars = 20 - 25%
One star = 25% -