The Effect of Analysts' Forecasts on Stock Market Returns: A Composite Multifactor Approach

50 Pages Posted: 3 Jun 2007 Last revised: 25 Mar 2009

See all articles by Stefano Bonini

Stefano Bonini

Stevens Institute of Technology - School of Business

Vincenzo Capizzi

University of Piemonte Orientale - Department of Economics and Business Studies; SDA Bocconi School of Management

Alessandro Paolo Luigi Cipollini

Deutsche Bank, Fixed Income Research; Università Degli Studi del Sacro Cuore di Milano - Laboratory of Statistics; University of Milan-Bicocca - Department of Mathematics and Applications

Fabrizio Erbetta

University of Piemonte Orientale

Date Written: March 31, 2007

Abstract

Stock returns forecasting is one of the major objectives of financial analysts. Equity Analysts' forecasts, on the other side, are one of the major sources of information used by less informed investors in their asset allocation decisions. Therefore, analysing which major drivers affect time series of stock returns could allow to shed light over the price revelation process in capital markets. In this paper we propose a model aimed at predicting stock market by combining both macroeconomic and microeconomic factors. We first develop a standard APT approach with multiple macroeconomic factors as regressors. We then integrate the model by explicitly including a metric for intrinsic equity value, basing upon a proxy derived by the weighted average of Stock Market Consensus Forecasts by equity analysts. Third, we complete the model by imposing an ARMA specification for the error term, which allows identifying stock returns' stationarity moving over time. The resulting model shows both a strong fitting capability when tested in the in-sample period and a good predictive capability when applied to an out-of-sample period of monthly Italian stock market returns. In particular, we employed specific estimation procedures based upon recently developed statistics aimed at testing for both factors' equal predicting power and forecast encompassing. As a major empirical finding, our model suggests that the information conveyed by analysts' forecasts is indeed a factor in determining future stock prices, even if there is the possibility that the information transferred could be biased.

Keywords: asset pricing models, forecasting stock returns, APT, autoregressive models, cost of equity

JEL Classification: C32, C53, G12

Suggested Citation

Bonini, Stefano and Capizzi, Vincenzo and Cipollini, Alessandro Paolo Luigi and Erbetta, Fabrizio, The Effect of Analysts' Forecasts on Stock Market Returns: A Composite Multifactor Approach (March 31, 2007). Available at SSRN: https://ssrn.com/abstract=990047 or http://dx.doi.org/10.2139/ssrn.990047

Stefano Bonini (Contact Author)

Stevens Institute of Technology - School of Business ( email )

Hoboken, NJ 07030
United States

Vincenzo Capizzi

University of Piemonte Orientale - Department of Economics and Business Studies ( email )

Via E. Perrone, 18
Novara, NO 28100
Italy
+390321375438 (Phone)
+390321375405 (Fax)

HOME PAGE: http://www.uniupo.it

SDA Bocconi School of Management ( email )

Via F. Bocconi, 8
Milan, Milan 20136
Italy
+390258365966 (Phone)
+390258366893 (Fax)

HOME PAGE: http://www.sdabocconi.it

Alessandro Paolo Luigi Cipollini

Deutsche Bank, Fixed Income Research ( email )

Winchester House
Great Winchester Street, 1
London EC2N 2DB
United Kingdom

Università Degli Studi del Sacro Cuore di Milano - Laboratory of Statistics ( email )

Via Necchi 9
Milano, MI 20121
Italy

HOME PAGE: http://www.unicatt.it

University of Milan-Bicocca - Department of Mathematics and Applications ( email )

Piazza dell'Ateneo Nuovo, 1
Milano, MI 20126
Italy
+390264485728 (Phone)

HOME PAGE: http://www.matapp.unimib.it

Fabrizio Erbetta

University of Piemonte Orientale ( email )

Rectory, via Duomo
Novara, Novara 6-13100
Italy

Register to save articles to
your library

Register

Paper statistics

Downloads
592
Abstract Views
2,627
rank
44,107
PlumX Metrics