2011 Housing Planning Study - Technical Report

Home

programming code itself cannot be affected by these parameter changes and users need not worry about damaging the model software. 2. Customizable Parameters: The more experienced user may wish to change entire sets of parameters such as household income, population, interest rates, or new construction. For this set of operations, entire sets of data, which combine starting levels, growth rates and growth curves are available to the user to change as needed. This procedure requires some brief training, but provides substantial latitude for developing model solutions to fit client needs. 3. Model Changes or Additions: Users with more knowledge of programming may wish to make substantial changes to the model design or add capacity or functionality to the model itself. Most users will want to contact SMS for this kind of work. For those who wish to try it themselves, unlocked versions of the model are available to Consortium members. Unlike previous models, the new Hawaii Housing Model provides a more comprehensive forecasting ability. It is possible, for instance, to use the model to estimate what the next housing price run-up will look like. Although the model is not designed to predict exactly what will happen and when, it shows a rough approximation of what supply and demand forces do to the housing market. The new model also comes with a caveat. Like all forecasting models, future projections depend heavily on the past behavior of the key data series -- the Hawaii housing market. Since the early eighties, Hawaii's housing market has witnessed three price run-ups, interspersed with adjustment periods. It is quite likely, then, that any manipulation of the model parameters may change the shape and character of the next priced run up, but will not eliminate them as features of the market trend. There are no data to suggest that a run-up will not occur in the future, or that the next run-up will not be followed by an adjustment period. Technical Specifications The Hawaii Housing Model was developed using Microsoft ExcelTM. The decision was made in 1997 by Consortium members to forego migrating to more complex modeling languages in order to assure that the model could be operated by a broad range of planners using commonly available hardware and software. Although this imposes some limitations on functionality, it seems worth the price. The model is available to all members of the Consortium and requires a standard PC with at least 2.0 gigs of RAM and about 30 Megs of storage space. Of course the user would need a reasonably recent version of Excel. The model was built up from a set of custom built user-defined functions (macros) in Excel. This allows for more complexity and options in the calculations. The specifics of the calculations will not be described here, but the calculations simulate the effects of inputs on supply and demand, prices, and affordability. The model also simulates the conditions that cause a price run-up, like the one that occurred from 2003-2006, and the effects of the run-up. Hawaii Housing Planning Study,2011: Technical Report Page 14 0 SMS, Inc. November,2011