Time-series adjustments and market conditions are required by Fannie Mae and Freddie Mac.

A recent FHFA research paper revealed that as few as 10% of appraisals had been even attempting a time adjustment, although in some markets, where the value was not “high enough,” the time adjustment came in handy to get closer to the needed number.

Cynical, yes.  True, it appears so.

Appraisers have had little or no responsible education/training in the topic of market analysis.  In fact, the concept of market analysis is relegated to something you do after you have picked some comps.  (See page 31 of The Appraisal of Real Estate.)

Now that time adjustments are required, suddenly experts are jumping out with advice of all kinds.  One article (based on artificial intelligence) suggested several different ways to get a time adjustment.  Each of these ways will produce a different number.  Hmmm.

You would think that this topic, price indexing and time-series analysis would have been well developed by economists over the years.

It has.

So why are amateurs suddenly providing clever ways?  Who is right?  A clever appraiser or AI newbie?  Or an experienced, practicing economist/statistician?  Hmmm.

Is it possible that after dozens of peer-reviewed journal articles, financial papers, and investment guides – that the methodology might be well developed, cleanly refined, and responsible?

We believe so.

And what is that method?  It is called time series analysis.  It includes simple, common-sense guidelines:

  1. Include only the right data (kinda like USPAP suggests).
  2. Scatter graph this data set (to connect with the brain).
  3. Draw and calculate a simple regression trend line.
  4. Apply the line slope to each comp date ($/day).

This can be absolutely applied to a 12 month period, as the GSEs now require for the neighborhood section of the form.  It can be visually adjusted for time back for the actual adjustments, if a longer or shorter period is relevant.  It can be visually modified if a change in trend is detected.

Also, any outliers, which do not seem to fit – can be dealt with.

That’s it.

The graph is useful for the appraiser/analyst, and also documents the appraisal report.

Better yet, time-series analysis MPI (Market Price Indexing) ©, forms the basis of modern EBV (Evidence Based Valuation)©, as taught in the Valuemetrics.info signature CE class Stats, Graphs, and Data Science1.  We also provide free and low-cost webinars on the topic.

Our sole goal is to help prepare the appraisal professional for tomorrow’s clients, not yesterday’s.