The Nested PIGLOG Model: An Application to U.S. Food Demand

A new demand system is introduced, the Nested PIGLOG model, nesting thirteen other demand systems including five that are also new. This new model and its nested special cases are applied to models of U.S. food demand that include food-at-home (FAH), food-away-from-home (FAFH), and alcoholic beverages. Although nested tests and out-of-sample forecasting performance favor generalizing models to a certain degree, statistically insignificant improvements to in-sample-fit and even poorer out-of-sample forecast accuracy undermine further generalizations. Based on a subset of preferred models, FAFH is found to be price and income elastic compared to FAH which is price and income inelastic.     

Robustness issues in multilevel regression analysis

A multilevel problem concerns a population with a hierarchical structure. A sample from such a population can be described as a multistage sample. First, a sample of higher level units is drawn (e.g. schools or organizations), and next a sample of the sub‐units from the available units (e.g. pupils in schools or employees in organizations). In such samples, the individual observations are in general not completely independent. Multilevel analysis software accounts for this dependence and in recent years these programs have been widely accepted. Two problems that occur in the practice of multilevel modeling will be discussed. The first problem is the choice of the sample sizes at the different levels. What are sufficient sample sizes for accurate estimation? The second problem is the normality assumption of the level‐2 error distribution. When one wants to conduct tests of significance, the errors need to be normally distributed. What happens when this is not the case? In this paper, simulation studies are used to answer both questions. With respect to the first question, the results show that a small sample size at level two (meaning a sample of 50 or less) leads to biased estimates of the second‐level standard errors. The answer to the second question is that only the standard errors for the random effects at the second level are highly inaccurate if the distributional assumptions concerning the level‐2 errors are not fulfilled. Robust standard errors turn out to be more reliable than the asymptotic standard errors based on maximum likelihood.     

Blocked-Error-R 2: A Conceptually Improved Definition of the Proportion of Explained Variance in Models Containing Loops or Correlated Residuals

Bentler and Raykov (2000, Journal of Applied Psychology 85: 125–131), and Jöreskog (1999a, http://www.ssicentral.com/lisrel/column3.htm, 1999b http://www.ssicentral. com/lisrel/column5.htm) proposed procedures for calculating R ² for dependent variables involved in loops or possessing correlated errors. This article demonstrates that Bentler and Raykov’s procedure can not be routinely interpreted as a “proportion” of explained variance, while Jöreskog’s reduced-form calculation is unnecessarily restrictive. The new blocked-error-R ² (beR ²) uses a minimal hypothetical causal intervention to resolve the variance-partitioning ambiguities created by loops and correlated errors. Hayduk (1996) discussed how stabilising feedback models – models capable of counteracting external perturbations – can result in an acceptable error variance which exceeds the variance of the dependent variable to which that error is attached. For variables included within loops, whether stabilising or not, beR ² provides the same value as Hayduk’s (1996) loop-adjusted-R ². For variables not involved in loops and not displaying correlated residuals, beR ² reports the same value as the traditional regression R ². Thus, beR ² provides a conceptualisation of the proportion of explained variance that spans both recursive and nonrecursive structural equation models. A procedure for calculating beR ² in any SEM program is provided.     

Behavior in one-shot traveler’s dilemma games: model and experiments with advice

We report results of one-shot traveler’s dilemma game experiments to test the predictions of a model of introspection. The model describes a noisy out-of-equilibrium process by which players reach a decision of what to do in one-shot games. To test the robustness of the model and to compare it to other models of introspection without noise, we introduce non-binding advice. Advice has the effect of coordinating all players’ beliefs onto a common strategy. Experimentally, advice is implemented by asking subjects who participated in a repeated traveler’s dilemma game to recommend an action to subjects playing one-shot games with identical parameters. In contrast to observations, models based on best-response dynamics would predict lower claims than the advised. We show that our model’s predictions with and without advice are consistent with the data.      

Observable Consequences of Trading Structure Differences: On the Use of Variance Ratios in Microstructure Studies

This paper examines the variance ratio tests in studies of transitory volatility and concludes that the variance ratio is an appropriate test of trading structure differences only under certain assumptions regarding the evolution of underlying stock prices and the autocorrelation structure of returns. This result raises caution as to the interpretation of results bases upon the 24-hour variance ratio methodologies in studies of transitory volatility and trading structure effects. A numerical example indicates that errors in inferences can be severe.     

Capitalized intangibles and financial analysts

We examine whether firms that capitalize a higher proportion of their underlying intangible assets have higher analyst following, lower dispersion of analysts’ earnings forecasts and more accurate earnings forecasts relative to firms that capitalize a lower proportion. Under Australian generally accepted accounting principles, capitalization of intangible assets has become increasingly ‘routine’ since the late 1980s. It is predicted that this experience leads Australian analysts to expect firms with relatively more certain intangible investments to signal this fact by capitalizing intangible assets. Our results are consistent with this. We find that capitalization of intangible assets is associated with higher analyst following and lower absolute earnings forecast error for firms with a stock of underlying intangible assets. Our tests suggest a weaker association between capitalization and lower earnings forecast dispersion. We conclude that there are benefits for analysts, for management to have the option to capitalize intangible assets. These findings suggest that IAS 38 Intangible Assets and AASB 138 Intangible Assets reduce the usefulness of financial statements.     

Conditional Persistence of Earnings Components and Accounting Anomalies

We suggest that the failure of investors to distinguish between an earnings component's autocorrelation coefficient (unconditional persistence) and the marginal contribution of that component's persistence to the persistence of earnings (conditional persistence) provides a partial explanation of post‐earnings‐announcement drift, post‐revenue‐announcement drift, and the accrual anomaly. When the conditional persistence of revenue surprises is high (low) relative to its unconditional persistence, both the post‐earnings‐announcement drift and the post‐revenue‐announcement drift are high (low), because investors’ under‐reaction to revenues and earnings is stronger when the persistence of revenue surprises is more strongly associated with the persistence of earnings surprises. Also, the mispricing of accruals decreases substantially when the conditional persistence of accruals is high relative to its unconditional persistence, because investors’ over‐reaction to accruals is mitigated when the persistence of accruals is indeed more strongly associated with the persistence of earnings. Our findings also suggest that financial analysts’ failure to distinguish between unconditional and conditional persistence of revenues and accruals results in more biased revenue and earnings predictions.     

Does Accrual Management Impair the Performance of Earnings‐Based Valuation Models?

This study examines how accrual manipulations affect firm valuation in the years surrounding the passage of the Sarbanes‐Oxley Act (SOX). We compare the absolute percentage pricing errors of RIM and DCF valuation models for a group of US firms suspected to have engaged in accrual manipulations to avoid a small loss or a small earnings decline vs. ‘Normal’ firms matched on industry, year and size. We find that RIM can better estimate intrinsic value than DCF for the matched Normal firms in the pre‐SOX period, but not so for accrual manipulators, and that SOX mitigates the harmful effect of accrual manipulations, completely eliminating the difference in RIM's accuracy advantage over DCF between Normal firms and accrual manipulators. As a further analysis, we redefine Suspect firms as real‐activity manipulators and find a significant across‐group difference in accuracy wedge in both sample periods, implying that SOX has prompted firms to favor real‐activity manipulations over accrual manipulations.     

Productivity Dispersion and Measurement Error

Several reasons have been put forward to explain the high dispersion of productivity across establishments: quality of management, different input usage and market distortions, to name but a few. Although it is acknowledged that a sizable portion of productivity dispersion may also be due to measurement error, little research has been devoted to identifying how much they contribute. We outline a novel procedure for identifying the role of measurement error in explaining the empirical dispersion of productivity across establishments. The starting point of our framework is the errors-in-variable model consisting of a measurement equation and a structural equation for latent productivity. We estimate the variance of the measurement error and subsequently estimate the variance of the latent productivity variable, which is not contaminated by measurement error. Using Norwegian data on the manufacture of food products, we find that about one percent of the measured dispersion stems from measurement error.     

Mobile diaries – Benchmark against metered measurements: An empirical investigation

Researchers seeking to study the relationships between consumers' communications, attitudes, and behaviors could benefit from monitoring consumers over time, across multiple locations and channels, and in a way that reflects consumers' subjective perceptions. Diaries on smartphones (mobile diaries) can be used as a research tool for such purposes. A mobile diary is a self-report instrument whereby people use their mobile handset to repeatedly report experiences of interest. Mobile diaries are increasingly used in psychology, geography, medicine, and commercial marketing. Yet they have rarely been used for quantitative marketing research, and were not benchmarked against best-practice metrics in marketing.In this study, we aim to set the ground for using mobile diaries in quantitative marketing research. We first lay out the theoretical infrastructure for the usage of mobile diaries, and describe possible respondent reporting concerns, including concerns related to non-reporting, reporting over time, and concerns stemming from individual-level heterogeneity.We demonstrate the potential of mobile diaries, as well as the importance of the various concerns, using a benchmark test case in the context of primetime TV viewing. Our benchmark uses a sample of respondents with both mobile diary viewing reports and Nielsen People Meter (NPM) records. Our analysis reveals that averaging across all conditions, 47.4%–64.7% of the NPM records are reported by the diary. The major sources for mismatch are random time periods without alarms, short viewings, and periodic reporting inactivity (pulsing). Concerns such as a decrease in reporting rates over time (e.g., fatigue), smartphone ownership, and demographic variation across individuals have relatively small effects on reporting likelihood. Analyzing the cases in which diary reports do not have a matching NPM record, we find many of them can be attributed to out-of-home viewing and viewing on non-metered devices. This finding demonstrates how mobile diaries can complement metered measurements. Overall, aggregate diary-based ratings have a 0.90 correlation with NPM ratings.We discuss implications for designing and using mobile diary studies in marketing.