Page 8 - Regression Guideline for AMC
P. 8
Steps in
Tes6ng Mul6ple Regression Model in Real Estate Appraising
4. Running a preliminary analysis on the selected proper6es using what is called bivariate regression. This type of analysis examines each property characteris6c individually to determine if it is associated with the last known sales prices of the selected proper6es.
5. Running a mul6ple regression model using the smaller subset of selected property characteris6cs to determine which of these reliably predicts differences in property values in the market. Based on the final regression model, es6mated values for the proper6es are obtained.
6. Selec6ng a small subset of comparable proper6es (3 to 9) that best match the subject property.
7. Adjus6ng the values of the comparable proper6es to the subject property using the values for the property characteris6cs determined by the regression to affect overall value to obtain a final es6mated value for the subject property.
8
