nach oben

Review of Accounting Studies

Erschienen in:

31.01.2020

Do excessively volatile forecasts impact investors?

verfasst von: Russell Lundholm, Rafael Rogo

Erschienen in: Review of Accounting Studies | Ausgabe 2/2020

Einloggen

Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.

search-config

KI-gestützte Suche

Aus

Abstract

There is a logical bound on the time-series variability of analyst forecasts; when variability exceeds this bound it must be caused by something besides statistically rational forecasting. We document occurrences of excessively volatile analyst forecasts and show that they influence investment performance. Comparing trading rules based on forecasts that are excessively volatile and those that are not, we find the returns to investing based on the former are significantly lower, with higher daily volatility, and a lower Sharpe ratio. We also show that returns to trading based on excessively volatile forecasts underperform the most when there is little news arriving and when the news that does arrive is relatively neutral. In this region, it is hardest to argue that analysts are unwittingly overreacting to news; instead, they appear to be intentionally making extreme forecasts to curry favor with management or to differentiate themselves from other analysts.

Vorheriger Artikel Discretionary disclosure and manager horizon: evidence from patenting

Nächster Artikel The market’s reaction to changes in relative performance rankings

Sie haben noch keine Lizenz? Dann Informieren Sie sich jetzt über unsere Produkte:

Springer Professional "Wirtschaft+Technik"

Online-Abonnement

Mit Springer Professional "Wirtschaft+Technik" erhalten Sie Zugriff auf:

über 102.000 Bücher
über 537 Zeitschriften

aus folgenden Fachgebieten:

Automobil + Motoren
Bauwesen + Immobilien
Business IT + Informatik
Elektrotechnik + Elektronik
Energie + Nachhaltigkeit
Finance + Banking
Management + Führung
Marketing + Vertrieb
Maschinenbau + Werkstoffe
Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

Jetzt informieren

Springer Professional "Wirtschaft"

Online-Abonnement

Mit Springer Professional "Wirtschaft" erhalten Sie Zugriff auf:

über 67.000 Bücher
über 340 Zeitschriften

aus folgenden Fachgebieten:

Bauwesen + Immobilien
Business IT + Informatik
Finance + Banking
Management + Führung
Marketing + Vertrieb
Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

Jetzt informieren

Starting with Shiller (1981), it has been observed that prices can be expressed as forecasts of future cash flows and that the variance of the underlying cash flows imposes a limit on the amount of statistically rational variation in prices. The subsequent literature produced a number of different approaches to estimating the variance of future cash flows and in mapping out the implied variation in prices, with the general conclusion that aggregate market indices have violated the implied variance bound. See Gilles and LeRoy 1991 for a review.

We label observations that, based on our estimates, violate the variance bound as “violations,” but we could label these observations as simply having “unusually large variance” with no change in our conclusions regarding the impact on investors.

Note that a strategy that simply goes long on upward revisions and short on downward revisions is insensitive to the extremity of the forecast revisions, the very thing we want to study. But to connect our work to the literature, we also examine a variation on our strategy that is very close to this simple long/short strategy.

The inequality follows from V(X) = E[V(X|Y)] + V[E(X|Y)], as found by DeGroot (1975, p. 183). Because V[E(X|Y)] is strictly positive for any nondegenerate joint distribution of X and Y, the inequality is strict.

In an untabulated analysis, we estimate our measures ignoring this 20.8% of the sample and find very similar results.

Equations six and seven of Pagan and Schwert (1990) give the approximately normal test statistic t = √8∗(τ/√2ν), where 8 is the number of observations in each variance estimate, τ is the difference in variance estimates, and ν = γ₀ + 2Σ⁴_j=1γ_j\( \left(1-\frac{j}{5}\right) \), where γ_j is the autocovariance function for lags 0 to 4. We exclude observations with t values outside the [−1.64, 1.64] range.

We base our trading rule on earnings forecasts rather than buy or sell recommendations, because it isn’t clear exactly what recommendations are a forecast of and recommendation changes occur much less frequently than forecast revisions.

In an untabulated analysis, we find that our results are very similar if the benchmark portfolio is matched on size, book-to-market, and the prior year return, as described by Daniel et al. (1997). We prefer the simpler size-matched benchmark, because size-oriented ETFs and mutual funds are readily available investment vehicles.

In untabulated results, we repeat our entire analysis with step-sizes of 0.10 and 0.30 and find very similar results. The trade-off between different step-sizes is between the strategy’s sensitivity to the revision versus the amount of time the strategy is constrained at one of the limits. We chose step-sizes of 0.20 because, when combined with quintile sorts, steps of 0.20 times the quintile rank allow the strategy to reach either the long or short limit without being constrained at the limit very often, as seen in Table 2.

This strategy is similar to those reported by Graham and Harvey (1996) and Chance and Helmer (2001), who investigate the performance of professional market timer recommendations. The recommendations take the form of a mix of investment between cash and the market index, with the weights constrained on the unit interval.

Our trading periods start each quarter but stay open for a year, resulting in an overlap the trading rules. In an untabulated analysis, we repeat our analysis using only the fourth quarter and get very similar results.

Annualizing the daily Sharpe ratio by simply multiplying by √252 has many hidden assumptions, as discussed by Lo (2002). In particular, it understates the true Sharpe ratio when there is positive serial correlation in the return series, as is most certainly the case for the returns to our dynamic trading strategies, because of the weight on the benchmark.

The average cumulative weight following a downward revision is −0.037 for the nonviolators and − 0.324 for violators (untabulated).

The annual returns in excess of the size-matched portfolio (either 3.0% for the nonviolating observations or 3.6% for the violating observations) are lower than the annual post-revision returns reported by Gleason and Lee, but this is likely due to the very different periods of the two studies. In an untabulated analysis, we can replicate the Gleason and Lee results very closely on their 1993–1998 sample period.

In untabulated analysis, we estimate the differences in returns, standard deviations, and Sharpe ratios, after controlling for period, firm, and analyst fixed effects. The results mirror those reported in Table 3. While it might be tempting to estimate period*firm fixed effects, 78% of the firm-quarters have either no violations or every analyst violates, and so no contrast would be available; consequently, the results would be confounded by selection. In section six, we present a variation on this idea that allows us to separate the selection from the relative performance of trading on violating or nonviolating forecasts.

An excerpt from Ravenpack’s Users Guide explains that the ESS is “A granular score between 0 and 100 that represents the news sentiment for a given entity by measuring various proxies sampled from the news. The score is determined by systematically matching stories typically categorized by financial experts as having short-term positive or negative financial or economic impact. The strength of the score is derived from a collection of surveys where financial experts rated entity-specific events as conveying positive or negative sentiment and to what degree. Their ratings are encapsulated in an algorithm that generates a score ranging from 0–100 where 50 indicates neutral sentiment, values above 50 indicate positive sentiment, and values below 50 show negative sentiment. ESS probes many different sentiment proxies typically reported in financial news and categorized by RavenPack. The algorithm produces a score for more than 2,000 types of business, economic, and geopolitical events, ranging from earnings announcements to terrorist attacks.”

Lundholm and Rogo (2016) found a significant contemporaneous correlation between the frequency of violations in a year and the two period variables, VIX and sentiment. In his commentary on that paper, Lee (2016) asks whether this relation is predictive; the results in Table 5 suggest it is not.

An even more refined strategy would combine the results of Chen et al. (2015) with our results and trade in the direction of a pseudo-violator’s revision if it is sufficiently negative. We leave the exploration of this strategy for future research.

Barber, B., & Lyon, J. (1997). Detecting long-run abnormal stock returns: The empirical power and specification of test statistics. Journal of Financial Economics, 43(3), 341–372.CrossRef

Barth, M. E., & Hutton, A. P. (2004). Analyst earnings forecast revisions and the pricing of accruals. Review of Accounting Studies, 9(1), 59–96.CrossRef

Brown, N. C., Wei, K. D., & Wermers, R. (2014). Analyst recommendations, mutual fund herding, and overreaction in stock prices. Management Science, 60(1), 1–20.CrossRef

Chan, L. K., Jegadeesh, N., & Lakonishok, J. (1996). Momentum strategies. The Journal of Finance, 51(5), 1681–1713.CrossRef

Chance, D., & Helmer, M. (2001). The performance of professional market timers: Daily evidence from executed strategies. Journal of Financial Economics, 62, 377–411.CrossRef

Chen, Q., & Jiang, W. (2006). Analysts’ weighting of private and public information. Review of Financial Studies, 19, 1–37.CrossRef

Chen, Q., Francis, J., & Jiang, W. (2005). Investor learning about analyst predictive ability. Journal of Accounting and Economics, 39(1), 3–24.CrossRef

Chen, X., Cheng, Q., & Lo, K. (2010). On the relationship between analyst reports and corporate disclosures: Exploring the roles of information discovery and interpretation. Journal of Accounting and Economics, 49(3), 206–226.CrossRef

Chen, P., G. Narayanamoorthy, T. Sougiannis, and H. Zhou. 2015. Analyst underreaction and the post-forecast revision drift. Available at SSRN: https://ssrn.com/abstract=2578757 or https://doi.org/10.2139/ssrn.2578757

Clement, M., & Tse, S. (2005). Financial analyst characteristics and herding behavior in forecasting. The Journal of Finance, 60, 307–341.CrossRef

Daniel, K., Grinblatt, M., Titman, S., & Wermers, R. (1997). Measuring mutual fund performance with characteristic-based benchmarks. The Journal of Finance, 52(3), 1035–1058.CrossRef

DeGroot, M. (1975). Probability and statistics. Reading: Addison-Wesley Publishing Co., Inc.

Gilles, C., & LeRoy, S. (1991). Econometric aspects of the variance-bounds tests: A survey. The Review of Financial Studies, 4, 753–791.CrossRef

Givoly, D., & Lakonishok, J. (1979). The information content of financial analysts’ forecasts of earnings: Some evidence on semi-strong inefficiency. Journal of Accounting and Economics, 1(3), 165–185.CrossRef

Gleason, C., & Lee, C. (2003). Analyst forecast revisions and market Price discovery. The Accounting Review, 78, 193–225.CrossRef

Graham, J., & Harvey, C. (1996). Market timing ability and volatility implied in investment newsletters’ asset allocation recommendations. Journal of Financial Economics, 42, 397–421.CrossRef

Griffin, P. (1976). Competitive information in the stock market: An empirical study of earnings, dividends, and analyst’ forecasts. Journal of Finance, 31, 631–650.

Hughes, J., Liu, J., & Su, W. (2008). On the relation between predictable market returns and predictable analyst forecast errors. Review of Accounting Studies, 13(2), 266–291.CrossRef

Ivković, Z., & Jegadeesh, N. (2004). The timing and value of forecast and recommendation revisions. Journal of Financial Economics, 73(3), 433–463.CrossRef

Kahneman, D., & Tversky, A. (1973). On the psychology of prediction. Psychological Review, 80, 237–251.CrossRef

Lee, C. (2016). Commentary on: “Do analysts forecasts vary too much?”. Journal of Financial Reporting, 1(1), 127–129.CrossRef

Livnat, J., & Zhang, Y. (2012). Information interpretation or information discovery: Which role of analysts do investors value more? Review of Accounting Studies, 17(3), 612–641.CrossRef

Lo, A. (2002). The statistics of Sharpe ratios. Financial Analysts Journal, 58(4), 36–52.CrossRef

Lundholm, R., & Rogo, R. (2016). Do analyst forecasts vary too much? Journal of Financial Reporting, 1(1), 101–123.CrossRef

Malmendier, U., & Shanthikumar, D. (2014). Do security analysts speak in two tongues? The Review of Financial Studies, 27(5), 1287–1322.CrossRef

Pagan, A., & Schwert, G. (1990). Testing for covariance Stationarity in stock market data. Economic Letters, 33, 165–170.CrossRef

Prendergast, C. and Stole, L., 1996. Impetuous youngsters and jaded old-timers: Acquiring a reputation for learning. Journal of political Economy, 104(6), 1105-1134.

Richardson, S., Teoh, S., & Wysocki, P. (2004). The walk-down to beatable analyst forecasts: The role of equity issuance and insider trading incentives. Contemporary Accounting Research, 21, 885–924.CrossRef

Shiller, R. (1981). Do stock prices move too much to by justified by subsequent changes in dividends? American Economic Review, 71, 421–436.

Shroff, P., Venkataraman, R., & Xin, B. (2014). Timeliness of analysts’ forecasts: The information content of delayed forecasts. Contemporary Accounting Research, 31(1), 202–229.CrossRef

So, E. (2013). A new approach to predicting analyst forecast errors: Do investors overweight analyst forecasts? Journal of Financial Economics, 108(3), 615–640.CrossRef

Stickel, S. (1991). Common stock returns surrounding earnings forecast revisions: More puzzling evidence. The Accounting Review, 66(2), 402–416.

Zhang, X. (2006a). Information uncertainty and analyst forecast behavior. Contemporary Accounting Research, 23, 565–590.CrossRef

Zhang, X. (2006b). Information uncertainty and stock returns. The Journal of Finance, 61, 105–137.CrossRef

Titel: Do excessively volatile forecasts impact investors?
verfasst von: Russell Lundholm
Rafael Rogo
Publikationsdatum: 31.01.2020
Verlag: Springer US
Erschienen in: Review of Accounting Studies / Ausgabe 2/2020
Print ISSN: 1380-6653
Elektronische ISSN: 1573-7136
DOI: https://doi.org/10.1007/s11142-019-09522-y

Springer Professional

Abstract

Bitte loggen Sie sich ein, um Zugang zu Ihrer Lizenz zu erhalten.

Sie haben noch keine Lizenz? Dann Informieren Sie sich jetzt über unsere Produkte:

Springer Professional "Wirtschaft+Technik"

Springer Professional "Wirtschaft"

Weitere Artikel der Ausgabe 2/2020

Sleeping with the enemy: should investment banks be allowed to engage in prop trading?

The changing implications of research and development expenditures for future profitability

Use of independent valuation specialists in valuing employee stock options: evidence from IPOs

Investor relations and IPO performance

The market’s reaction to changes in relative performance rankings

Audit partner identification and audit quality