This product is a recording of the Appraisal Institute webinar that was conducted on October 1, 2014.
Once the Appraisal Institute confirms your payment, you will receive an e-mail containing a link to the webinar recording.
Please note: The interactive features of the webinar (e.g. poll/survey questions and the final evaluation) are NOT available in the recording. Appraisal Institute continuing education credit is NOT available for viewing the recording.
Regression Modeling: Why Bad Results Happen To Good Appraisers (And What To Do About Them)
Date: October 1, 2014
Duration: 2 hours
Regression has become a mainstream tool for valuation and consulting assignments. While putting data into a software program and getting an answer is easy, understanding how each data component impacts the overall model design and results is much more complex. This webinar focuses on the use of regression modeling for real estate valuation, and the results of such modeling.
In this webinar, you will:
How many independent variables are enough?
Is a bigger sample always better?
What is the meaning of the significance of a variable?
What is the impact of the variability of the dependent variable on the results?
What are the impacts of data errors and outliers on the results?
How can regression residuals be used to improve the model?
What should you do when there isn't a linear relationship between the dependent and independent variables?
Kenneth M. Lusht,, Ph.D., MAI, SRA