Loss Function Assumptions in Rational Expectations Tests on Financial Analysts' Earnings Forecasts

By Sudipta Basu, Stanimir Markov. 2004

146 citations recorded by the source corpus.

Subfields: Accounting, Finance, General Decision Sciences

View Paper: https://www.semanticscholar.org/paper/36eb2eb08853ebeedf6dc5229139502697e32d5f

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Accounting

1. Sean D. Campbell
2. Steven A. Sharpe

In Finance

1. Sean D. Campbell
2. Steven A. Sharpe
3. Jorge Caiado

In General Decision Sciences

1. Sean D. Campbell
2. Steven A. Sharpe

Citations in the corpus

• Anchoring Bias in Consensus Forecasts and Its Effect on Market Prices (2007, methodology): “As an alternative, Basu and Markov (2004) suggest the use of a linear loss function, which is implemented vis-à-vis median regression, based on the notion that a 3% forecast error may only be three times more costly than a 1% error, rather than nine times as costly.”
• The Financial Analyst Forecasting Literature: A Taxonomy with Suggestions for Further Research (2008, unknown): “Basu and Markov (2004) Archival, I/B/E/S, 1985-2001.”
• Linear valuation without OLS: the Theil-Sen estimation approach (2014, unknown): “2008), but at least two papers do, Basu and Markov (2004) and Hughes et al.”
• Analysts’ Forecasts: What Do We Know after Decades of Work? (2011, unknown): “ Gu and Wu (2003) and Basu and Markov (2004): These papers question analysts’ loss functions implied by prior work that uses ordinary least squares models to link forecast errors and various measures (implying a quadratic loss function) by proposing that analysts’ might prefer to minimize the absol”
• Biased Forecasts or Biased Earnings? The Role of Reported Earnings in Explaining Apparent Bias and Over/Underreaction in Analysts' Earnings Forecasts (2003, unknown): “For example, after Basu and Markov (2003) replaced the quadratic assumption in their standard OLS regression with a linear loss function assuming that analysts minimize absolute forecast errors, they found little evidence to support analyst inefficiency.”