US Existing Home Sales: November 2014 Preview

Existing home sales in the US are expected to decline slightly to an annual pace of 5.25 million units in tomorrow’s update for November, according to The Capital Spectator’s median point forecast for several econometric estimates. The projection represents a marginal decrease from October’s 5.26 million units (seasonally adjusted annual rate).

Three estimates based on recent surveys of economists point to a modestly deeper decline for existing home sales in November relative to The Capital Spectator’s median projection.

Here’s a closer look at the numbers, followed by brief definitions of the methodologies behind The Capital Spectator’s forecasts that are used to calculate the median estimate:

VAR-6: A vector autoregression model that analyzes six economic series to project existing home sales: housing starts, new single-family home sales, newly issued permits for residential construction, the monthly supply of homes for sale, monthly national average of 30-year conventional fixed-rate mortgage, and US private-sector payrolls.  VAR analyzes the interdependent relationships of these series with existing home sales through history. The forecasts are run in R using the “vars” package.

ARIMA: An autoregressive integrated moving average model that analyzes the historical record of existing home sales in R via the “forecast” package.

ES: An exponential smoothing model that analyzes the historical record of existing home sales in R via the “forecast” package.

TRI: A model that’s based on combining point forecasts, along with the upper and lower prediction intervals (at the 95% confidence level), via a technique known as triangular distributions. The basic procedure: 1) run a Monte Carlo simulation on the combined forecasts and generate 1 million data points on each forecast series to estimate a triangular distribution; 2) take random samples from each of the simulated data sets and use the expected value with the highest frequency as the prediction. The forecast combinations are drawn from the following projections:’s consensus forecast data and the predictions generated by the models above (VAR-6, ARIMA, and ES). The forecasts are run in R with the “triangle” package.