Zillow expects existing home sales for November to fall 1.7 percent to a seasonally adjusted annual rate (SAAR) of 5.17 million units, down from 5.26 million units in October.  But the models we use and the assumptions that underlie them are not always as definitive as we would like.

Below, we provide greater detail about the assumptions that went into our November 2014 forecast.

Zillow’s existing home sales forecasts uses two models:

• A “structural” model that estimates home sales as a function of other economic data.
• A “historical” model that estimates home sales as a function of past movements in existing home sales as well as recent pending home sales data.

More information about the models and adjustment made this month can be found in the Methodology section below.

According to the historical model, when pending home sales increase, existing home sales tend to rise in the subsequent months.  But when existing home sales increase one month, they are expected to fall in coming months.

For November, our historical model suggests that existing home sales will fall by 1.7 percent to a seasonally adjusted annual rate (SAAR) of 5.17 million units. The decline can be attributed to to a fall in pending home sales and a rise in existing home sales in October.

The structural model typically provides greater insight than the historical model into the broader economic forces that drive existing home sales.  However, depending on the month, it can rely heavily on forecasts of explanatory variables.  In particular, our forecasts of the homeownership rate and homeowner vacancy rate – which we must forecast anywhere from one to four months ahead as they are only released on a quarterly schedule – can have important effects on the model estimates.  For November, we must forecast these variables two months ahead. Median family income is also an important driver, and since this was last made available in June, we must forecast it five months ahead.

A very naïve forecast is done assuming the variables listed above will not change from the last time they were made available.  These assumptions predict a slight increase in existing home sales of 0.1 percent, to 5.26 million units (SAAR). However, these variables have exhibited trends that suggest the naïve forecast is not very realistic.

Median family income has shown a strong tendency to move with the Consumer Price Index (CPI) for Services. Thus, for the following scenarios, median family income is assumed to increase at a rate consistent with services CPI.

The homeownership rate has been steadily declining.  Holding all else constant, if it continues its decline through November at its month-over-month average over the past year, existing home sales could fall by 1.3 percent, to 5.19 million units (SAAR).

The homeowner vacancy rate has shown slight increases over the past two months, in contrast to its steady decline throughout the recovery.  If nothing else were to change but the homeowner vacancy rate rose in October at a rate consistent with Zillow’s for-sale inventory series, then sales could fall by 1.1 percent to 5.20 million units (SAAR).

If both of these predictions were to come true, then existing home sales would decline 1.4 percent to the forecasted 5.18 million units (SAAR).

Our model is less sensitive to other assumptions, including the following:

• We estimate that the median sales price of existing homes sold in November will increase to $212,500 – the result of a best-fit ARIMA model, an ARIMA(1,1,2) in this instance.
• We estimate that median family income will increase to $66,900 in November, from $66,300 in June, in line with growth in the Consumer Price Index (CPI) for Services.
• We estimate that the number of households will increase to 116.1 million in November, the result of a best-fit ARIMA model – an ARIMA(1,2,2), in this instance.

Methodology

Structural Model

The structural model estimates existing home sales as a function of key macroeconomic fundamentals – including the contemporaneous homeownership rate, the homeowner vacancy rate, home sales prices, interest rates, family incomes and the total number of households.

The adjustment made to this month’s forecasting model was to estimate monthly changes in existing home sales as a function of monthly differences in the fundamental variables.  The original model estimated new home sales as a function of levels of some fundamental variables and yearly percent changes on others.  This adjustment improved the mean average percent error by about 45 percent.

Historical Model

The historical model uses past existing home sales data as well as past values of pending home sales. We model the number of existing home sales in each month from April 2001 to June 2014 as an ARIMAX(2,1,2) process, with three lags of pending home sales as external regressors from one to three months prior. This specification minimized the Akaike and Bayesian information criteria among various combinations of one to three lags tested.