PKF Hospitality Research Forecast Accuracy Remains High

ATLANTA—PKF Hospitality Research (PKF-HR), a CBRE Company, released the results of its latest assessment of the accuracy of its Hotel Horizons forecasts for the U.S. lodging industry. This is the fourth periodic review conducted by PKF-HR since 2005 and is an important and critical component of the firm’s forecasting efforts, according to the company.

“Once again we are pleased to report that the accuracy of our forecasts remains high,” said R. Mark Woodworth, senior managing director of PKF-HR. “Accuracy on the national level continues to far exceed generally accepted benchmarks. At the local level, accuracy is strong for the most part. However, there are markets where the level of precision will be improved.

“Periodic accuracy assessments are extremely important to us. Internally, they help us improve our models and forecasting procedures. Externally, these tests of accuracy allow our clients to calibrate the decisions they are making based on our forecasts,” added Woodworth.

PKF-HR’s Hotel Horizons is a series of hotel forecast reports that analyze the historical and expected performance of U.S. lodging markets. Driven by a series of econometric forecasting models, the Hotel Horizons reports cover five years of supply, demand, occupancy, ADR and RevPAR for 59 major U.S. markets, as well as six national chain-scale segments and six national location categories. Within each market forecast, separate estimates are prepared for upper-price and lower-price hotels.

“To test the accuracy of our Hotel Horizons forecasts, we compared our projections to both actual lodging performance data and forecasts that could have been produced simultaneously using an intuitive approach,” said John B. “Jack” Corgel, PhD., the Robert C. Baker professor of real estate at the Cornell University School of Hotel Administration and senior advisor to PKF-HR. “Comparisons such as these provide assessments of our current methodologies and may lead to potential alternative approaches.”

PKF-HR tested both the long-term and short-term accuracy of its array of forecasts during the period from 2010 through 2014. Various statistical measures, such as mean absolute error and Thiel’s U statistics, were used to evaluate five key metrics: supply, demand, occupancy, ADR and RevPAR.

Consistent with historical findings, PKF-HR’s national forecasts were found to be the most accurate. “No matter what you are forecasting, the larger the target, the greater the accuracy. We see this with our forecasts of national chain-scale and location performance,” Woodworth said.

Relatively higher levels of accuracy also were found in PKF-HR’s short-term vs. long-term forecasts as well as demand vs. supply projections. “Since 2010, we have incorporated variables into our supply forecast models to simulate the conservative lending environment. Given the strong growth in demand, and changes in real ADR that have occurred, traditional economics dictates that more hotel rooms should have been built the past few years,” said Corgel.

“At the local level, our forecasts largely proved to err on the side of caution.  In 41 of the 50 MSA’s we examined, actual performance exceeded our forecast levels of performance,” said Jamie Lane, senior economist at PKF-HR. “For example, we forecast Nashville to achieve the second highest RevPAR growth in the nation for 2014.  Even this robust forecast was exceeded.” Only 50 of the Hotel Horizons markets were analyzed for accuracy because the remaining nine forecast reports were initiated after 2010.

“At the local level, the judgmental input provided by PKF Consulting’s and CBRE Hotel’s regional professionals plays an important role. For example, non-econometric factors such as Hurricane Sandy and threats of a federal government shutdown had a significant impact on lodging performance of the New York/Long Island and Washington, DC, markets respectively,” Lane said.

The cyclical nature of the U.S. lodging industry also has a significant influence on the accuracy of the PKF-HR forecasts. “Forecast accuracy tends to decline around turning points in the hotel cycle and improves as time moves further from the turning point. Right now, we are in an ‘up phase,’ when the precision of our forecasts and hotel budgets are most accurate,” said Woodworth.

“Most industry prognosticators are calling for two to three more years of growth for the U.S. lodging industry. This implies that our forecasts and hotel budgets will continue to be very accurate With accuracy comes a greater sense of assurance, and less risk. In this environment, all industry participants can proceed with their respective plans knowing with a high degree of certainly what the business environment will look like,” concluded Woodworth.