Recent presentations and articles on “Big Data and Appraisal” universally say that big changes in technology and analytics are affecting appraisal. What I don’t hear is: What can I do about it?
Will I be able to work in five years? Will our appraisal organizations be around? If there is a place for me, what do I need to do?
None of the articles show me what to do. The tell me: “be prepared;” “learn regression;” “use more advanced tech- and data-centric appraisal tools;” or “combine your traditional experience with more rigorous statistical analysis.” And many, many others just explain that if you just buy our product, all will be well. It is all easy. Easy. Just push the button and your appraisal are faster, better and wonderful. Well, if it were that easy, we could just hire non-licensed people to just push the button. Aw shucks, it’s an AVM. That’s what they do. In fact, it is dangerous for a USPAP appraiser to push the AVM button. What I read is that big data is a big deal — that will change everything. And I’d better be ready. I’d better be ready. Ok, I’m ready. But to do what? They are not telling me what!
So what do I do? What do we do? It appears that to survive, we must provide what is needed by future fate, not of steady past. It is the same, but different. What is the future? What is different?
It appears that our value is in three areas: 1) our knowledge of markets; 2) our ability to gather and improve data, and, 3) analyze the data into a value prediction. How can we improve our competence in each of these three areas — given the machines are gaining on us? Should we fight the machines, like we have fought AVMs and such in the past? Or should we partner up? The algorithm machines belong to no one. We can use them — or they can use them. Who takes charge?
So let’s look at how we can be in control of the three areas we can control: market knowledge, data wrangling, and computation analytics.
Market understanding is easily improved by the use of graphs, plots, and descriptives. Analytics software (like R)– specifically designed for visuals and summaries is readily available.
Data Wrangling includes identifying, cleaning, enhancing, and transforming data. These four elements change data into usable information. “If you get the comps right, it is hard to be far wrong!”
Computation means understanding and using reusable parts of the machine. These are called algorithms. Algorithms make models more accurate and reliable.
In sum – the ideal valuation of the future includes an SME (subject matter expert), with data competence, and computation skills. Computation means maximizing the brain-machine result.
The ultimate mission of this blog, and the Valuemetrics.info curriculum is to bring these three skills, and three types of scientific reasoning to the profession. The result is a truer and surer appraisal point value. But wait there is more! These methods enable additional products. Products and services best provided by SME’s. Experts like you. We will explain these additional appraiser income streams in future posts.
It is the goal of this blog, and our expanding curriculum — to match and exceed the needs of the market in the near future. It is survival. It is opportunity. And . . . it can be more fun.
July 5, 2017 @ 7:21 am
The charts, graphs, and spreadsheets that make use of the data that is available must be included in the report as an exhibit and be incorporated into the report narrative in order to have any meaning. The more complete understanding of the market’s “Big Picture” that the collated data provides to the appraiser must be shared with the reader/Client who has retained the appraiser’s services as an SME instead of trying to do the appraisal in-house and running the risk of obtaining a less than credible result. The appraiser owes the client the duty and obligation of bringing to the client the most complete and credible analysis of the property being appraised. Big Data is merely one aspect of the research and analysis involved in the appraisal.
July 5, 2017 @ 4:38 pm
Great article. Just today I had an appraiser friend email me about yet another new technology to help visually show neighborhood values. It seems technology is creeping up on appraisers and prodding our profession to do a better job understanding, explaining, and showing value visually. It certainly prods me to up my game and find ways to tell the story of value for my market.
July 5, 2017 @ 6:40 pm
Yes, the end goal is to create a product which is scientifically reproducible. A number of benefits evolve from reproducibility: 1) a reliability score can be calculated; 2) Subjective review is all but eliminated, replaced by an audit function; 3) fundamental (“intrinsic”) value can also be calculated, to assist in collateral risk and investment decisioning; 4) real-time valuation (like updates) are feasible and practical. These and other products can be part of the future appraiser’s quiver of arrows.
In the Stats, Graphs, and Data Science1 class, we emphasize (for residential) visually seeing neighborhood values and predictor variables, exactly in order to identify the ideal CMS (Competitive Market Segment). The open-source software we recommend, really makes these brain-machine functions fun, as well as making for a better product.
George Dell
July 10, 2017 @ 11:30 pm
Thank you George. What I’m doing is learning everything I can about statistics, but I have also hedged my bets by adding several other non appraisal services to my appraisal company. Like with driver-less cars, it is hard to predict when exactly the technology will take over, how quickly it will happen, and how complete the technology will take over. My guess is that there will be a need for appraisers to value complex properties for some time, but I’m not putting all my eggs in the the appraisal basket.