The methods used in long‐range prediction include causal models and regression on leading variables, but one method which is generally worth considering is the fitting of trend lines. It is the object of this paper to present simple techniques for fitting these lines.

Introduction A product may be called new for two reasons: either because it is a novelty in the fullest sense of the word, intended to cater for a hitherto latent demand, or merely because it is a new brand, which will have to compete with a number of other brands in an established market.