Forecasting branded and generic pharmaceuticals

Abstract

We forecast UK pharmaceutical time series before and after the time of patent expiry. This is a critical point in the respective lifecycle as a generic form of the product is introduced to the market, while the branded form is still available for prescription. Forecasting the number of dispensed units of branded and generic forms of pharmaceuticals is increasingly important due to their huge market value and the limited number of new ‘blockbuster’ branded drugs, as well as the imposed cost to national healthcare systems like the NHS. In this paper, eleven methods are used to forecast drug time series including Diffusion Models (Bass model & RPDM), ARIMA, Exponential smoothing (Simple and Holt), naive and regression methods. ARIMA and Holt produce accurate short term (annual) forecasts for branded and generic drugs respectively, while for the more strategic horizons of 2-5 year ahead, Naive with drift provides the most accurate forecasts.