TEA.16.2

Description

One of the methodological considerations for proper life care plan valuation is the choice of forecast for future increases in the cost of medical goods and services, as discussed in Trevino (2013). The present work examines forecasts based on the inflation rate for medical services component of the consumer price index (CPI). Unlike broad CPI, published consensus forecasts of medical CPI are unavailable, but simple forecast methods are available that can incorporate historical data. A simple forecast method is to use a “moving average” of current and past inflation rates when forecasting future rates. To identify suitable time windows for applying moving averages, forecast performance statistics are available including mean squared error and mean absolute error. Among moving averages, those based on a relatively short time window tend to perform better than those based on longer time windows in application to medical CPI inflation over the years 1948-2017. An opposite result holds when forecasting the gap between medical CPI and general CPI inflation rates, at long horizons: longer time windows of 10 to 20 years provide better forecasts of the long-horizon inflation gap.