Log-Optimal Economic Evaluation of Probability Forecasts

Johnstone, David

doi:10.1111/j.1467-985x.2011.01011.x

Cited by 13 publications

(5 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“… For equivalent CAPM equations derived under power and log utility functions, see Satchell () and Johnstone (). …”

mentioning

confidence: 99%

The Effect of Information on Uncertainty and the Cost of Capital

Johnstone

2015

Contemporary Accting Res

View full text Add to dashboard Cite

It is widely held that better financial reporting makes investors more confident in their predictions of future cash flows and reduces their required risk premia. The logic is that more information leads necessarily to more certainty, and hence lower subjective estimates of firm "beta" or covariance with other firms. This is misleading on both counts. Bayesian logic shows that the best available information can often leave decision makers less certain about future events. And for those cases where information indeed brings great certainty, conventional mean-variance asset-pricing models imply that more certain estimates of future cash payoffs can sometimes bring a higher cost of capital. This occurs when new or better information leads to sufficiently reduced expected firm payoffs. To properly understand the effect of signal quality on the cost of capital, it is essential to think of what that information says, rather than considering merely its "precision," or how strongly it says what it says. L'incidence de l'information sur l'incertitude et le coût des capitaux R ESUM E L'id ee qu'une meilleure information financi ere accroît la confiance des investisseurs dans leurs pr edictions des flux de tr esorerie futurs et r eduit les primes de risque qu'ils exigent est largement r epandue. La logique veut qu'un suppl ement d'information engendre n ecessairement davantage de certitude, et donc des estimations subjectives plus faibles du bêta de la soci et e ou de la covariance avec d'autres soci et es. Ces suppositions sont l'une et l'autre trompeuses. La logique bay esienne montre que la meilleure information disponible peut, dans bien des cas, diminuer la certitude des d ecideurs en ce qui a trait aux ev enements futurs. Et lorsque l'information procure de fait une grande certitude, les mod eles traditionnels d' evaluation des actifs moyenne-variance supposent que des estimations plus certaines des rentr ees de fonds futures peuvent parfois entraîner a la hausse le coût des capitaux. Cette situation se pr esente lorsque de l'information nouvelle ou de meilleure qualit e m ene a des gains pr evus suffisamment r eduits. Pour bien comprendre l'incidence de la qualit e du signal sur le coût des capitaux, il est essentiel de se demander ce que livre cette information plutôt que d'en juger par sa « pr ecision » ou par la fermet e des affirmations qu'elle contient.

show abstract

“… For equivalent CAPM equations derived under power and log utility functions, see Satchell () and Johnstone (). …”

mentioning

confidence: 99%

The Effect of Information on Uncertainty and the Cost of Capital

Johnstone

2015

Contemporary Accting Res

View full text Add to dashboard Cite

show abstract

“…The results to follow would not be possible without first establishing a “payoffs version” of the log CAPM. The model is derived as a generalized CRRA (power utility) model based on the derivations in Johnstone (2012) and Satchell (2012).…”

Section: Log Utility Resultsmentioning

confidence: 99%

Accounting information, disclosure, and expected utility: Do investors really abhor uncertainty?

Johnstone

2021

Business Fin & Account

Self Cite

View full text Add to dashboard Cite

Investors are said to “abhor uncertainty,” but if there were no uncertainty they could earn only the risk‐free rate. A fundamental result in the analytical accounting literature shows that investors buying into a CARA‐normal CAPM market pay lower asset prices, gain higher ex‐ante expected returns, and obtain higher expected utility, when the market payoff has higher variance. New investors obtain similar “welfare” gains from risk under a log/power utility CAPM. These results do not imply that investors “abhor information.” To realize investors' ex‐ante expectations, the subjective probability distributions representing market expectations must be accurate. Greater payoff risk can add to investors' expected utility, but higher ex‐post(realized) utility comes from better information and more accurate ex‐ante expectations. An important implication for accounting is that greater disclosure can have the simultaneous effects of (i) exposing more fully or perceptibly firms' payoff uncertainty, thereby increasing new investors' expected utility, and (ii) improving market estimates of firms' payoff parameters (means, variances, covariances), thereby giving investors a better chance of realizing their expectations. Paradoxically, better information can be valuable to new investors by exposing more fully and more accurately the risk in firms' business operations and results–new investors maximizing expected utility want both more risk and better information.

show abstract

“…This information may be particularly helpful when one intends to include mean-variance optimization methods in the analysis. Another possible extension is to assess alternative sets of combination weights of models and strategies, see also Johnstone (2012). Further, analysis of the behavior of stocks within an industry is relevant for more detailed portfolio analysis.…”

Section: Discussionmentioning

confidence: 99%

Forecast density combinations of dynamic models and data driven portfolio strategies

Baştürk

Borowska

Grassi

et al. 2019

Journal of Econometrics

View full text Add to dashboard Cite

A dynamic asset-allocation model is specified in probabilistic terms as a combination of return distributions resulting from multiple pairs of dynamic models and portfolio strategies based on momentum patterns in US industry returns. The nonlinear state space representation of the model allows efficient and robust simulation-based Bayesian inference using a novel non-linear filter. Combination weights can be crosscorrelated and correlated over time using feedback mechanisms. Diagnostic analysis gives insight into model and strategy misspecification. Empirical results show that a smaller flexible model-strategy combination performs better in terms of expected return and risk than a larger basic model-strategy combination. Dynamic patterns in combination weights and diagnostic learning provide useful signals for improved modelling and policy, in particular, from a risk-management perspective. * We thank two anonymous referees and the guest-editors, Sylvia Kaufman and Sylvia Frühwirth-Schnatter, for their valuable comments on earlier versions of this paper which led us to prepare an extended and substantially revised version of Baştürk, Grassi, Hoogerheide and Van Dijk (2016b).1 These industry returns are constructed by equally weighting all stock returns in the specific industry, which is similar to Moskowitz and Grinblatt (1999) The data are retrieved from

show abstract

Log-Optimal Economic Evaluation of Probability Forecasts

Cited by 13 publications

References 59 publications

The Effect of Information on Uncertainty and the Cost of Capital

The Effect of Information on Uncertainty and the Cost of Capital

Accounting information, disclosure, and expected utility: Do investors really abhor uncertainty?

Forecast density combinations of dynamic models and data driven portfolio strategies

Contact Info

Product

Resources

About