Forecasting earnings growth

Abstract: Estimating future earnings growth is an essential part of company valuation. Often, only historical earnings growth is used in the estimation process, despite several studies showing little correlation between past and future earnings growth. The aim of this study is to investigate whether a multivariate forecast model incorporating payout ratio can generate more accurate earnings growth forecasts compared to two univariate models which solely incorporate historical growth. These models are used to estimate earnings growth of Swedish companies listed on Nasdaq Stockholm for the time period 2010-2014. The results show that the multivariate model explains observed earnings growth and thus has the ability to forecast earnings growth. Although the multivariate model to some extent generates smaller forecast errors than the univariate models, it cannot be statistically proven whether the model is superior to the univariate models.