Hotel Horizons^SM Reports

How We Forecast

The Method

Wikipedia defines ‘forecasting’ as:

“the process of estimation in unknown situations. Prediction is a similar, but more general term, and usually refers to estimation of time series, cross-sectional or longitudinal data. In more recent years, forecasting has evolved into the practice of demand planning in every day business forecasting for manufacturing companies. The discipline of demand planning, also sometimes referred to as supply chain forecasting, embraces both statistical forecasting and consensus process...(emphasis added)”

Consistent with this definition, PKF Hospitality Research prepares forecasts of the hotel markets in the U.S. based on generally accepted econometric procedures and sound judgment regarding fundamental relationships between the economic and behavioral market indicators and hotel financial performance, relationships that PKF has tracked for over 70 years.

The Econometric Component

Econometric forecasting represents one of the most sophisticated approaches to gaining insights about future economic activity. Unlike some forecasting methods used in business practice, the models that underlie econometric forecasts rely on historical relationships, similar to statistical correlations, between economic variables. The data for measuring these variables come from actual market transactions involving individuals and firms interacting in the economy. Moreover, these economic variables allowed to enter econometric models have conceptual linkages to economic theory.

Positive Features of an Econometric Model

1. The variables included in the models follow from economic theory.
2. The relationships between economic variables are estimated with advanced statistical methods.
3. The forecasts developed with econometric models are more objective than forecasts purely based on judgmental approaches.

Gaining insight into the future of complicated economic environments requires the introduction of multidimensional forecasting models. Thus, several equations often need to be identified and estimated to model complex economic conditions, as in the national economy. Multi-equation models have considerable appeal for economic forecasting because they explicitly recognize the interdependence of relationships commonly encountered in business and economics. The best examples of this type of model are demand and supply models, in which prices of goods are set by the interaction of buyers and sellers in the market. Thus, price appears as a variable in both the demand and supply equations.

The Equations

The econometric forecasting models, because they represent an entire sector of the national and MSA economies, fall into the category of multi-equation, demand and supply models. These models have a general structure as defined below, but vary in their form for particular market applications:

1. Demand (the number of rooms occupied) is the dependent variable in this equation, and equals either Gross Domestic (Metropolitan) Product, Real Personal Income, or Total Employment, which serve as the main independent variable, along with the lagged changes in any of these variables and the lagged demand from the prior year (different numbers of lags are used for independent variables based statistical significance).
2. Supply Change (change in the number of rooms available) is the dependent variable, equals ADR and Occupancy, which serve as the main independent variables along with the change in supply from the prior period (different numbers of lags are used for independent variables based statistical significance).
3. ADR (Real) is the dependent variable, equals Occupancy, which is the main independent variable along with ADR from the prior period (different numbers of lags are used for independent variables based statistical significance).

These equations are estimated with ordinary least squares in a non-simultaneous fashion using data from Smith Travel Research and Moodys Economy.com dating back to the late 1980s. The parameters (coefficients on each variable) then are used to forecast demand, supply change, and ADR by multiplying the parameters with Moodys Economy.com forecast of the economic variables and relevant previously estimated values (lagged variables). Two additional calculations are made with these results as follows:

1. The Supply Change is added to the previous period number of rooms available to produce a rooms available level in future periods.
2. Number of rooms sold is divided by number of rooms available to obtain occupancy percent in each future period.

The Judgmental Component

The econometric model predicts future room supply in small increments (e.g., 100 rooms per quarter). In reality, rooms enter the market in large blocks (e.g., 1000 rooms) as new hotels are placed in service. When it becomes apparent that a new hotel (s) will be placed in service within the next 18 months, the modeled supply will be manually adjusted to account for the opening of the new hotel(s). The reverse is also true when it become apparent that a hotel(s) will be taken out of service (e.g., demolished or converted to an alternative use).

Finally, a committee of hotel experts form PKF Hospitality Research performs a through review of each model prediction. In the spirit of forecasting described above in the Wikipedia definition, this committee modifies predictions from the model in situation when there is compelling evidence that factors have come into play in a market which the model could not possibly foresee. In the extreme case, a Katrina-style event causes the Committee’s forecast to differ noticeable from the model prediction. In most instances, however, the committee either defers to the model prediction or makes modest adjustments.

For more information, please contact our PKF-HR Publications Department at (866) 842-8754, or claude.vargo@pkfc.com.