Inference about parameters in empirical linear Bayes estimation problems

Abstract

It is shown how one may construct tests and confidence regions for the unknown structural parameters in empirical linear Bayes estimation problems. The case of the collateral units having varying “designs” (i.e. regressor and covariance matrices) may be treated under the assumption that the design variables independently follow a common statistical law. The results are of an asymptotic nature.     

The New Narrative: Applying narratology to the shaping of futures outputs

Both scenario development and design practices incorporate elements of storytelling, but this use remains undertheorised. This paper will draw upon literary theory, film theory and science fiction criticism to develop an analytical model of narrative structure and rhetorics which speaks to the concerns of scenario developers and designers when engaged in shaping the final outputs or deliverables of a futures project.After highlighting the differing role of telos in art and futures and defining the metacategory of “narratives of futurity”, this paper then defines the terms “story”, “narrative”, “narrator” and “world” in the literary context. It then shows how those concepts map onto futures practice, before going into detail regarding the variety of narrative strategies available across a range of different forms and media, and the qualitative effects that they can reproduce in audiences. There follows the construction of a 2 × 2 matrix based on the critical concepts of narrative mode and narrative logic, within which narratives of futurity might be usefully catalogued and compared, and from which certain broad conclusions may be reached as regards the relation between choice of medium and rhetorical effect. The implications of this analysis are explored in detail.     

Connected and autonomous vehicle injury loss events: Potential risk and actuarial considerations for primary insurers

The introduction of connected and autonomous vehicles (CAVs) to the road transport ecosystem will change the manner of collisions. CAVs are expected to optimize the safety of road users and the wider environment, while alleviating traffic congestion and maximizing occupant comfort. The net result is a reduction in the frequency of motor vehicle collisions, and a reduction in the number of injuries currently seen as “preventable.” A changing risk ecosystem will introduce new challenges and opportunities for primary insurers. Prior studies have highlighted the economic benefit provided by reductions in the frequency of hazardous events. This economic benefit, however, will be offset by the economic detriment incurred by emerging risks and the increased scrutiny placed on existing risks. We posit four plausible scenarios detailing how an introduction of these technologies could result in a larger relative rate of injury claims currently characterized as tail‐risk events. In such a scenario, the culmination of these losses will present as a second “hump” in actuarial loss models. We discuss how CAV risk factors and traffic dynamics may combine to make a second “hump” a plausible reality, and discuss a number of opportunities that may arise for primary insurers from a changing road environment.     

Navigating in the Midst of More Uncertainty and Risk

With the steady increase in the variety and scale of uncertainties and risks, the challenges for today's executives have become ever more complex and daunting. One powerful tool for navigating among different risks and uncertainties is scenario planning. From its early days of use within Shell, scenario planning has evolved in ways that make it better suited to the tasks of identifying, analyzing, and managing various financial risks across different industries. During the last ten years, Morgan Stanley has also been using scenario planning to gain a better understanding of key risks and uncertainties facing the financial services industry, ranging from the consequences of possible changes in the dollar to the emergence of hedge funds and the remarkable growth of China and India. In discussing the benefits of scenario planning, the authors note its potential to help management in a number of ways:

• ? By challenging conventional thinking and current assumptions about its industry and world;
• ? By identifying key signals or potential direction changes ahead of time, which is especially important when lead times to invest, hedge, or change assets are limiting factors;
• ? By identifying and assessing the value of strategic or “real” options—options to invest in new opportunities or limit downside risks that may suddenly open up or disappear, and that man‐ agement must be prepared to “exercise” quickly and decisively;
• ? By reinforcing the recognition that value added comes not just from better strategic thinking and planning, but from the role of risk management in helping companies take advantage of new opportunities;
• ? By encouraging more cross‐divisional and firm‐wide conversations about strategic choices and options, thereby creating a shared understanding of and greater consensus about chosen strategies; and
• ? By forcing them to go beyond the limits of typical three‐to‐five year forecasting limitations to think hard about longer‐term strategic choices.

     

Forecasting market risk of portfolios: copula-Markov switching multifractal approach

This paper proposes a new methodology for modeling and forecasting market risks of portfolios. It is based on a combination of copula functions and Markov switching multifractal (MSM) processes. We assess the performance of the copula-MSM model by computing the value at risk of a portfolio composed of the NASDAQ composite index and the S&P 500. Using the likelihood ratio (LR) test by Christoffersen [1998. “Evaluating Interval Forecasts.” International Economic Review 39: 841–862], the GMM duration-based test by Candelon et al. [2011. “Backtesting Value at Risk: A GMM Duration-based Test.” Journal of Financial Econometrics 9: 314–343] and the superior predictive ability (SPA) test by Hansen [2005. “A Test for Superior Predictive Ability.” Journal of Business and Economic Statistics 23, 365–380] we evaluate the predictive ability of the copula-MSM model and compare it to other common approaches such as historical simulation, variance–covariance, RiskMetrics, copula-GARCH and constant conditional correlation GARCH (CCC-GARCH) models. We find that the copula-MSM model is more robust, provides the best fit and outperforms the other models in terms of forecasting accuracy and VaR prediction.     

Identification of Maximal Affine Term Structure Models

Building on Duffie and Kan (1996) , we propose a new representation of affine models in which the state vector comprises infinitesimal maturity yields and their quadratic covariations. Because these variables possess unambiguous economic interpretations, they generate a representation that is globally identifiable. Further, this representation has more identifiable parameters than the “maximal” model of Dai and Singleton (2000) . We implement this new representation for select three‐factor models and find that model‐independent estimates for the state vector can be estimated directly from yield curve data, which present advantages for the estimation and interpretation of multifactor models.     

Advertising dynamics and competitive advantage

Can advertising lead to a sustainable competitive advantage? To answer this question, we propose a dynamic model of advertising competition where firms repeatedly advertise, compete in the product market, and make entry as well as exit decisions. Within this dynamic framework, we study two different models of advertising: in the first model, advertising influences the goodwill consumers extend toward a firm (“goodwill advertising”), whereas in the second model it influences the share of consumers who are aware of the firm (“awareness advertising”). We show that asymmetries may arise and persist under goodwill as well as awareness advertising. The basis for a strategic advantage, however, differs greatly in the two models of advertising. We show that tighter regulation or an outright ban of advertising may have anticompetitive effects and discuss how firms use advertising to deter and accommodate entry and induce exit in a dynamic setting.     

Assessing Consumer Fraud Risk in Insurance Claims

Abstract

We present an unsupervised learning method for classifying consumer insurance claims according to their suspiciousness of fraud versus nonfraud. The predictor variables contained within a claim file that are used in this analysis can be binary, ordinal categorical, or continuous variates. They are constructed such that the ordinal position of the response to the predictor variable bears a monotonic relationship with the fraud suspicion of the claim. Thus, although no individual variable is of itself assumed to be determinative of fraud, each of the individual variables gives a “hint” or indication as to the suspiciousness of fraud for the overall claim file. The presented method statistically concatenates the totality of these “hints” to make an overall assessment of the ranking of fraud risk for the claim files without using any a priori fraud-classified or -labeled subset of data. We first present a scoring method for the predictor variables that puts all the variables (whether binary “red flag indicators,” ordinal categorical variables with different categories of possible response values, or continuous variables) onto a common –1 to 1 scale for comparison and further use. This allows us to aggregate variables with disparate numbers of potential values. We next show how to concatenate the individual variables and obtain a measure of variable worth for fraud detection, and then how to obtain an overall holistic claim file suspicion value capable of being used to rank the claim files for determining which claims to pay and the order in which to investigate claims further for fraud. The proposed method provides three useful outputs not usually available with other unsupervised methods: (1) an ordinal measure of overall claim file fraud suspicion level, (2) a measure of the importance of each individual predictor variable in determining the overall suspicion levels of claims, and (3) a classification function capable of being applied to existing claims as well as new incoming claims. The overall claim file score is also available to be correlated with exogenous variables such as claimant demographics or highvolume physician or lawyer involvement. We illustrate that the incorporation of continuous variables in their continuous form helps classification and that the method has internal and external validity via empirical analysis of real data sets. A detailed application to automobile bodily injury fraud detection is presented.     

Time Variation in the Covariance between Stock Returns and Consumption Growth

The conditional covariance between aggregate stock returns and aggregate consumption growth varies substantially over time. When stock market wealth is high relative to consumption, both the conditional covariance and correlation are high. This pattern is consistent with the “composition effect,” where agents' consumption growth is more closely tied to stock returns when stock wealth is a larger share of total wealth. This variation can be used to test asset‐pricing models in which the price of consumption risk varies. After accounting for variations in this price, the relation between expected excess stock returns and the conditional covariance is negative.     

Management control of public and not-for-profit activities

Traditional approaches to management control usually fail for public and not-for-profit activities.¹ The type of control applicable to such activities depends on four criteria: are objectives unambiguous, outputs measurable, effects of interventions known, and is the activity repetitive? Depending on where activities stand with regard to these criteria, the control applicable corresponds to one of six different types: routine, expert, trial-and-error, intuitive, judgemental, or political control. The first three types can be represented by cybernetic models; the other three ask for more complex and less deterministic models. For these, a “political” and a “garbage-can” model are described. Key elements in the latter models are the values and the culture of the actors. As an example, the topology for management control is applied to the area of budgeting, covering regular budgeting as well as such techniques as PPBS, MBO, and ZBB and distinguishing between investment budgets, operations budgets for input centers, and operations budgets for input-output centers. Coming back to management control in general, the paper discusses the consequences of choosing the wrong model for a given management control situation: it distinguishes between “Type I” and “Type II” errors. It finally relates management control to organizational adaptation and suggests how to avoid control systems which prevent an organizational system from learning.     

A NOTE ON THE POSSIBILITY OF INCOME MANIPULATION THROUGH THE ABSORPTION OF FIXED OVERHEAD

In an absorption costing system manipulation of budgeted fixed manufacturing costs and “normal volume” can affect reported net income. This note outlines the relationships involved and how they are affected by the treatment of under or over-applied overhead.     

On optimal stratifications for multivariate distributions

Summary

In the present paper we study the problem of optimal stratifications for estimating the mean vector y of a given multivariate distribution F(x) with covariance matrix ζ both in cases of proportionate and of optimal (or generalized Neyman) allocations. It is noted that an “optimal stratification” is meant for one to make the covariance matrix of an unbiased estimator X for μ minimal, in the sense of semi-order defined below, in the symmetric matrix space. We show the existence of an optimal stratification and the necessary conditions for a stratification to be optimal. Besides we prove that an optimal stratification can be represented by a “hyperplane stratification” or a “quadratic hypersurface stratification” according to the proportionate or optimal (or generalized Neyman) allocation, and that the set of all optimal (or admissible) stratifications is a minimal complete class in the analogous sense of decision theory. Further we discuss the optimal stratification when a criterion based on a suitable real-valued function is adopted instead of the semi-order.     

Scenario development for corporate planning

The term “scenario” is familiar to those involved in forecasting, but too few people are aware of what exactly a scenario is, or how it can best be developed and applied. The author describes a method developed over several years in response to a need which most forecasting efforts have left unfulfilled. The method enables quantitative and qualitative forecasts to be combined in a manner which can be directly related to an organisation's planning and decision-making processes, and which permits the evaluation of a company's objectives and performance in the light of those forecasts. The analysis of an organisation's likely performance in given scenarios can, in turn, provide a basis for contingency planning.     

Determinants of asymmetric return comovements of gold and other financial assets

Using conditional time-varying copula models, we characterize the dependence structure of return comovements of gold and other financial assets (stocks, bonds, real estate and oil) during economic expansion and contraction regimes. We also investigate which key macroeconomic and non-macroeconomic variables significantly impact the asset return comovements using a two stage Markov Switching Stochastic Volatility (MSSV) framework. Our results show that the non-macro variables have significant influence on the return comovements. We find that gold is an inappropriate hedge against interest rate changes for real-estate and oil-based portfolios, while for bond portfolios, gold offers a good hedge against inflation uncertainty. We also provide evidence that the “flight to safety” phenomenon is due to the implied volatility of the stock market, rather than the observed stock market uncertainty. Finally, we forecast the asset return comovements and examine their economic significance. We show that a dynamic MSSV model which includes the macroeconomic and non-macroeconomic variables yields superior forecast of future asset return comovements when compared with a multivariate conditional covariance model.     

The Rise of Agency Capitalism and the Role of Shareholder Activists in Making It Work

The past 50 years have seen a fundamental change in the ownership of U.S. public companies, one in which the relatively small holdings of many individual shareholders have been supplanted by the large holdings of institutional investors, such as pension funds, mutual funds, and bank trust departments. Such large institutional investors are now said to own over 70% of the stock of the largest 1,000 U.S. public corporations; and in many of these companies, as the authors go on to note, “as few as two dozen institutional investors” own enough shares “to exert substantial influence, if not effective control.” But this reconcentration of ownership does not represent a complete solution to the “agency” problems arising from the “separation of ownership and control” that troubled Berle and Means, the relative powerlessness of shareholders in the face of a class of “professional” corporate managers who owned little if any stock. As the authors note, this shift from an era of “managerial capitalism” to one they identify as “agency capitalism” has come with a somewhat new and different set of “agency conflicts” and associated costs. The fact that most institutional investors hold highly diversified portfolios and compete (and are compensated) on the basis of “relative performance” provides them with little incentive to engage in the vigorous monitoring of corporate performance and investor activism that could address shortfalls in such performance. As a consequence, such large institutional investors—not to mention the large and growing body of indexers like Vanguard and BlackRock—are likely to appear “rationally apathetic” about corporate governance. But, as the authors also point out, there is a solution to this agency conflict—and to the corporate governance “vacuum” that has been said to result from the alleged apathy of well‐diversified (and indexed) institutional investors: the emergence of shareholder activists. The activist hedge funds and other specialized activists who have come on the scene during the last 15 or 20 years are now playing an important role in supporting this relatively new ownership structure. Instead of taking control positions, the activists “tee‐up” strategic business and financing choices that are then decided upon by the vote of institutional shareholders that are best characterized not as apathetic, but as rationally “reticent”; that is, they allow the activists, if not to do their talking for them, then to serve as a catalyst for the expression of institutional shareholder voice. The institutions are by no means rubber stamps for activists' proposals; in some cases voting for the activists' proposals, in many cases against them, the institutions function as the long‐term arbiters of whether such proposals should and will go forward. In the closing section of the article, the authors discuss a number of recent legal decisions that appear to recognize this relatively new role played by activists and the institutions that choose to support them (or not)—legal decisions that appear to confirm investors' competence and right to be entrusted with such authority over corporate decision‐making.     

Real Options Valuation: A Case Study of an E‐commerce Company

This paper presents a real options valuation model with original solutions to some issues that arise frequently when trying to apply these models to real‐life situations. The authors build on existing models by introducing an innovative and intuitive risk neutral adjustment that allows us to work with all the simulated paths. The problem of incorporating real options into each path is solved with a “nearest neighbors” technique, and uncertainty is simulated using a beta distribution that adapts better to company‐specific information. The model is then applied to a real life e‐commerce company to produce the following insights: the expanded present value is higher than the traditional present value; the presence of several real options make them interact so that their values are nonadditive; and part of the expanded present value is explained by the presence of “Jensen's inequality” that stems from the “convexity” between the value of each year's cash flow and the uncertain variables.     

Policy models in accounting: A critical commentary

The accounting profession, through its rule-making apparatus, employs a particular choice model for identifying and resolving policy issues. This model is referred to in this article as the “constitional” approach. It is of interest to inquire why accountants have adopted this paradigm and eschewed other policy models, such as the “legal” or “scientific” approaches, even though the latter have generated considerable advocacy within the recent history of the profession. This paper examines the interesting hypothesis that the choice of the “constitutional” approach is consistent with the Group Self-Interest Hypothesis. By forming an associative group the major accounting firms have a primary interest in achieving higher prices for the industry's products. These higher prices can be achieved in three ways: (1) restricting entry, (2) restraining supply and (3) increasing demand. The “constitutional” approach is more efficacious in achieving these results than are the proposed alternative policy models.     

Modeling the cross-section of stock returns using sensible models in a model pool

An increase in the number of asset pricing models intensifies model uncertainties in asset pricing. While a pure “model selection” (singling out a best model) can result in a loss of useful information, a full “model pooling” may increase the risk of including noisy information. We make a trade-off between the two methods and develop a new two-step trimming-then-pooling method to forecast the joint distributions of asset returns using a large pool of asset pricing models. Our method allows investors to focus on certain regions of the distributions. In the first step, we trim the uninformative models from a pool of candidates, and in the second step, we pool the forecasts of the surviving models. We find that our method significantly enhances portfolio performance and predicts downside risk precisely, and the improvements are mainly due to trimming. The pool of sensible models becomes larger when focusing on extreme events, responds rapidly to rising uncertainty, and reflects the magnitude of factor premiums. These findings provide new insights into asset pricing model evaluation.     

Modeling asymmetric comovements of asset returns

Existing time-varying covariance models usually impose strongrestrictions on how past shocks affect the forecasted covariancematrix. In this article we compare the restrictions imposedby the four most popular multivariate GARCH models, and introducea set of robust conditional moment tests to detect misspecification.We demonstrate that the choice of a multivariate volatilitymodel can lead to substantially different conclusions in anyapplication that involves forecasting dynamic covariance matrices(like estimating the optimal hedge ratio or deriving the riskminimizing portfolio). We therefore introduce a general modelwhich nests these four models and their natural 'asymmetric'extensions. The new model is applied to study the dynamic relationbetween large and small firm returns.