An empirical study of regression analysis as an analytical procedure*

Abstract. A newly issued AICPA auditing standard focuses attention on analytical procedures. Regression analysis has been shown to be a useful audit tool and is used to a limited extent in practice. This study compares a univariate regression-based decision rule with that of exponential smoothing. The effect on the regression-based decision method when additional input information is included to develop a multivariate model is also evaluated. Comparisons are accomplished by seeding various error patterns into the audit period data and evaluating the results of the various models. The results indicate that the regression-based decision model was at least as efficient and effective as the exponential smoothing-based model. Additional input information into the univariate regression model to develop a multivariate model did improve auditor decisions for some types of accounts but did not significantly affect the number of incorrect rejections and/or acceptances for other types. The multivariate model did improve the achieved precision of the univariate model but still did not reach desired levels. Résumé. Dans un Auditing Standards Procedures qu'il publiait récemment, l'AICPA se penche sur les procédés analytiques. L'analyse de régression s'est révélée un instrument de vérification utile et son emploi dans la pratique est modéré, Les auteurs comparent une règle de décision fondée sur une régression comportant une seule variable aléatoire avec celle du lissage exponentiel. L'incidence d'un supplément d'information à l'entrée sur la méthode de décision fondée sur la régression permet de mettre au point un modèle à plusieurs variables aléatoires, que les auteurs évaluent également. Les comparaisons sont réalisées en introduisant divers scénarios d'erreur dans les données de la période soumise à la vérification. Les résultats de l'étude indiquent que le modèle de décision fondé sur la régression est au moins aussi efficient et efficace que le modèle fondé sur le lissage exponentiel. L'introduction d'un supplément d'information dans le modèle de régression à une seule variable aléatoire de manière à créer un modèle à plusieurs variables aléatoires a de fait amélioré les décisions du vérificateur pour certains types de comptes, mais n'a pas eu d'incidence significative sur le nombre d'erreurs de première et de seconde espèces pour d'autres types de comptes. La performance du modèle à plusieurs variables aléatoires est de fait supérieure à celle du modèle à une seule variable aléatoire, sans toutefois permettre d'obtenir les niveaux de précision souhaités.     

Option Pricing Bounds in Discrete Time

Upper and lower bounds are derived for call options traded at discrete intervals. These bounds are independent of assumptions on the stock price distribution other than a restriction satisfied by the stock being “non-negative beta.” The development of the bounds relies on the single-price law and arbitrage arguments. Both single-period and multiperiod results are produced, and put option bounds follow by extension. The bounds exist as equilibrium values given a consensus on stock price distribution; they are also valid for empirical studies, being adjustable for dividends and commissions.     

Managerial Ownership and Accounting Conservatism

In this paper we examine the effect of managerial ownership on financial reporting conservatism. Separation of ownership and control gives rise to agency problems between managers and shareholders. Financial reporting conservatism is one potential mechanism to address these agency problems. We hypothesize that, as managerial ownership declines, the severity of agency problem increases, increasing the demand for conservatism. Consistent with our hypothesis, we find that conservatism as measured by the asymmetric timeliness of earnings declines with managerial ownership. The negative association between managerial ownership and asymmetric timeliness of earnings is robust to various controls, in particular, for the investment opportunity set. We thus provide evidence of a demand for conservatism from the firm's shareholders.     

RESIDENTIAL PHOTOVOLTAIC ENERGY SYSTEMS IN CALIFORNIA: THE EFFECT ON HOME SALES PRICES

Relatively little research exists estimating the marginal impacts of photovoltaic (PV) energy systems on home sale prices. Using a large data set of California homes that sold from 2000 through mid‐2009, we find strong evidence, despite a variety of robustness checks, that existing homes with PV systems sold for a premium over comparable homes without PV systems, implying a near full return on investment. Premiums for new homes are found to be considerably lower than those for existing homes, implying, potentially, a trade‐off between price and sales velocity. The results have significant implications for homeowners, builders, appraisers, lenders, and policymakers. (JEL R31, D12, C33)