Coherent Modeling and Forecasting of Mortality Patterns for Subpopulations Using Multiway Analysis of Compositions: An Application to Canadian Provinces and Territories

Mortality levels for subpopulations, such as countries in a region or provinces within a country, generally change in a similar fashion over time, as a result of common historical experiences in terms of health, culture, and economics. Forecasting mortality for such populations should consider the correlation between their mortality levels. In this perspective, we suggest using multilinear component techniques to identify a common time trend and then use it to forecast coherently the mortality of subpopulations. Moreover, this multiway approach is performed on life table deaths by referring to Compositional Data Analysis (CoDa) methodology. Compositional data are strictly positive values summing to a constant and represent part of a whole. Life table deaths are compositional by definition because they provide the age composition of deaths per year and sum to the life table radix. In bilinear models the use of life table deaths treated as compositions generally leads to less biased forecasts than other commonly used models by not assuming a constant rate of mortality improvement. As a consequence, an extension of this approach to multiway data is here presented. Specifically, a CoDa adaptation of the Tucker3 model is implemented for life table deaths arranged in three-dimensional arrays indexed by time, age, and population. The proposed procedure is used to forecast the mortality of Canadian provinces in a comparative study. The results show that the proposed model leads to coherent forecasts.     

Competition in Retail Petrol Markets

We study the relationship between prices and market structure in geographically isolated markets that are exposed to large demand shocks. The temporal variation in market size allows us to overcome the classical endogeneity bias in standard concentration‐performance regressions. We find evidence of local market power in petrol markets due to product differentiation. Additionally, the high margins that characterise concentrated markets dissipate quickly with the number of competitors. Ignoring market structure endogeneity leads to underestimating the effect of market concentration on prices between 55 and 70%.     

Determinants of pension funding and asset allocation decisions

Prior research suggests that the funding and asset allocation decisions for defined benefit pension plans may be based on tax, risk, and profitability factors. Much of the previous empirical work, however, suffers from statistical problems that may produce misleading or contradictory results. We employ a confirmatory factor analytic model to address the statistical problems plaguing pension research. Various competing hypotheses are tested simultaneously. Findings indicate that firms use pensions to offset business risk.An earlier version of this article was presented at the Financial Management Association Meetings held in Toronto, October 1993. Much of the work on this article was done while the authors were at the University of Texas-Arlington.     

Determinants of Equity Style

This paper provides a unifying empirical treatment of propositions explaining equity style cycles with a four-factor model that combines risk factors central to style theory. Tests on style autocorrelations and performance over the period January 1979–December 2004 generally affirm theoretical expectations. We employ cointegration methodology to analyze the stationarity of style covariances and dissect the diversification contributions of styles. We document style diversification gains but discover an asymmetry: value gains are derived from small company stocks while growth benefits come from large stocks. The asymmetry implies portfolios comprised of independent large growth and small value styles since the twin small growth and large value styles are cointegrated and redundant diversifiers. Performance tests show superior performance by the independent styles over the sample period, two equal intertemporal periods, and an extended 5-year period that directly contradicts a risk based explanation. The influence of institutional traders on style trends is also documented. Our findings affirm the predictions of behavioral models and provide more empirical evidence of superior performance unrelated to risk or fundamentals.

Global Property Market Diversification

This paper examines global diversification benefits provided by developed property markets over 1992–2007. We employ a cointegration methodology, invariant to pair-wise correlational instability plaguing MPT approaches, to investigate regional and country property market diversification benefits for U.S. domiciled global real estate investors. We show, theoretically and empirically, the cointegration procedure aptly identifies markets integrated by common trends that mitigate diversification potential. We show global property markets are interregionally independent but find intraregional market cointegration. A portfolio of markets independent of cointegrating relationships performs best during the period but is insufficiently diversified relative to a cointegrated portfolio. Independent country markets do account for the bulk of global property diversification gains but cointegrated markets, particularly from the North American and Asia Pacific regions, retain some diversifying qualities. We also show cointegrated markets converge toward benchmark characteristics, reducing their attraction as portfolio candidates.     

Optimal Vertical Arrangements When Resale Is Possible

This paper pinpoints optimal vertical arrangements in settingscharacterized by incomplete contracting and resale of an intermediateinput (a "widget"). In the Grossman-Hart-Moore property rightstheory, we conclude that sometimes strictly complementary assetsshould be owned separately to permit the emergence of a secondarymarket. In a richer model where the parties choose specificand nonspecific investments, vertical separation may also dominatejoint ownership. The article then examines the profitabilityof three integration forms when the proposed bargaining modelsubstitutes random-order values (e.g., the Shapley value). Theconclusions differ markedly from existing claims. (JEL C70,C78, D23, L42)     

Special issue: Qualitative and quantitative methods in tourism research

Quality & Quantity -     

Direct effects of money on aggregate demand: another look at the evidence

Now that a number of central banks are faced with short-term nominal interest rates close to or at the zero lower bound, there is a renewed interest in the long-running debate about whether or not changes in the stock of money have direct effects. In particular, do changes in money have additional effects on aggregate demand outside of those induced by changes in short-term nominal interest rates? This article revisits and reinterprets the empirical evidence based on single equation regressions which is quite mixed, with some results supporting and other results denying the existence of direct effects. We use a structural model with no direct effects of money to show that the finding of positive and statistically significant coefficients on real money growth can be misleading. The model generates data that, when used to estimate analogues of the empirical regressions, produce positive and statistically significant coefficients on real money growth, similar to those often found when using actual data. The problem is that single equation regressions leave out a set of variables, which in turn, give rise to an omitted variables bias in the estimated coefficients on real money growth. Hence, they are an unreliable guide to calibrate monetary policies, in general, including at the zero lower bound.