Consensus Forecasts of Corporate Earnings: Analysts' Forecasts and Time Series Methods

Abstract

An alternative to using a single forecasting method is to average the forecasts made by various methods. In this paper we examine empirically combinations of financial analysts' forecasts and forecasts from time series methods in order to predict corporate earnings per share. We conclude that, on average, the primary forecasting advantages of analysts over time series methods based on annual data appear to occur over short forecast horizons (less than a year). Neither analysts nor other time series methods substantially outperform a random walk prediction of no change when forecasts are made near the beginning of the fiscal year. For predictions in the first half of the fiscal year, there is evidence of forecasting benefits from combining time series and analysts' forecasts, especially if there are few analysts' forecasts.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 122 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.