Out-of-sample forecast tests robust to the choice of window size

This paper proposes new methodologies for evaluating out-of-sample forecastingperformance that are robust to the choice of the estimation window size. The methodologies involve evaluating the predictive ability of forecasting models over a wide rangeof window sizes. We show that the tests proposed in the literature may lack the powerto detect predictive ability and might be subject to data snooping across differentwindow sizes if used repeatedly. An empirical application shows the usefulness of themethodologies for evaluating exchange rate models' forecasting ability.

On the tractability of the piecewise-linear approximation for general discrete-choice network revenue management

The choice network revenue management (RM) model incorporates customer purchase behavioras customers purchasing products with certain probabilities that are a function of the offeredassortment of products, and is the appropriate model for airline and hotel network revenuemanagement, dynamic sales of bundles, and dynamic assortment optimization. The underlyingstochastic dynamic program is intractable and even its certainty-equivalence approximation, inthe form of a linear program called Choice Deterministic Linear Program (CDLP) is difficultto solve in most cases. The separation problem for CDLP is NP-complete for MNL with justtwo segments when their consideration sets overlap; the affine approximation of the dynamicprogram is NP-complete for even a single-segment MNL. This is in contrast to the independentclass(perfect-segmentation) case where even the piecewise-linear approximation has been shownto be tractable. In this paper we investigate the piecewise-linear approximation for network RMunder a general discrete-choice model of demand. We show that the gap between the CDLP andthe piecewise-linear bounds is within a factor of at most 2. We then show that the piecewiselinearapproximation is polynomially-time solvable for a fixed consideration set size, bringing itinto the realm of tractability for small consideration sets; small consideration sets are a reasonablemodeling tradeoff in many practical applications. Our solution relies on showing that forany discrete-choice model the separation problem for the linear program of the piecewise-linearapproximation can be solved exactly by a Lagrangian relaxation. We give modeling extensionsand show by numerical experiments the improvements from using piecewise-linear approximationfunctions.

A new heuristic method for solving spatially constrained forest planning problems based on mitigation of infeasibilities radiating outward from a forced choice

[Abstract]

Forest owners' acceptance of incentive based policy instruments in forest biodiversity conservation : a choice experiment based approach

Abstract

Reply to the letter to the editor "Freedom of choice".

Réponse au commentaire de: Gnädinger M. Freedom of choice. Swiss Med Wkly. 2012 Mar 22;142:0. doi:10.4414/smw.2012.13527. PMID: 22441992.

Intergenerational differences in cardiovascular risk factor levels in Switzerland

Background: There is little information regarding cardiovascular risk factor (CV RF) trends in Switzerland. We aimed at assessing generation differences in CV RFs by comparing CV RFs levels within selected age groups separated by a 20 year time lag. Design: Two population-based surveys. Methods: Data from the Monica (1984-1986) and Colaus (2004-2006) surveys were used. Analyses were stratified by sex and age groups (35-44, 45-54, 55-64 and 65-75 years). Results: No changes were found for BMI levels and status between surveys: in men, 26}3, 26}3, 27}4 and 27}4 kg/m2 for age groups 35-44, 45-54, 55-64 and 65-74, respectively, in MONICA, vs. 26}4, 26}4, 27}4 and 28}4 kg/m2 in COLAUS, p=NS, in women: 24}4, 26}4, 26}4 and 26}5 kg/m2 in MONICA, vs. 24}5, 25}5, 26}5 and 26}5 kg/m2 in COLAUS, p=NS. Similar results were found after adjusting for education. Smoking prevalence increased in men: 28, 30, 22 and 15% for age groups 35-44, 45-54, 55-64 and 65-74, respectively, in MONICA, vs. 35, 29, 28 and 21% in COLAUS. In women, changes differed according to age: 39, 26, 16 and 18%, in MONICA vs. 28, 30, 22 and 15% in COLAUS. Blood pressure decreased in the younger age groups and remained constant in the older ones: in men, systolic blood pressure was 129}15, 133}16, 138}18 and 143}21 mm Hg in MONICA, vs. 125}12, 129}15, 137}16 and 144}19 mm Hg in COLAUS, p<0.01. Similar findings were obtained after adjusting for education. Prevalence of hypertension increased, due to an increase in the prevalence of treated subjects, in men : 4, 8, 16 and 19% for age groups 35-44, 45-54, 55-64 and 65-74, respectively, in MONICA, vs. 5, 14, 31 and 46% in COLAUS, p<0.05; in women: 2, 10, 16, and 24% in MONICA, vs. 4, 12, 24, and 34% in COLAUS, p<0.05. This increase was stronger in men: 14, 17, 23 and 31% for age groups 35-44, 45-54, 55-64 and 65-74, respectively, in MONICA vs. 10, 21, 41 and 55% in COLAUS, p<0.01 and smaller in women: 6, 15, 24 and 44% in MONICA vs. 6, 16, 30 and 42% in COLAUS, p=NS. Similar findings were obtained after adjusting for education. Conclusion: With the exception of BMI, the newer Swiss generations appear to have a worse CV profile than the older generations. This is especially true regarding smoking and hypertension.

Choice of Optimal MembraneProcesses for Economical Treatment of Paper Machine Clear Filtrate

In this thesis membrane filtration of paper machnie clear filtrate was studied. The aim of the study was to find membrane processes which are able to produce economically water of sufficient purity from paper machine white water or its saveall clarified fractions for reuse in the paper machnie short circulation. Factors affecting membrane fouling in this application were also studied. The thesis gives an overview af experiments done on a laboratory and a pilot scale with several different membranes and membrane modules. The results were judged by the obtained flux, the fouling tendency and the permeate quality assessed with various chemical analyses. It was shown that membrane modules which used a turbulence promotor of some kind gave the highest fluexes. However, the results showed that the greater the reduction in the concentration polarisation layer caused by increased turbulence in the module, the smaller the reductions in measured substances. Out of the micro-, ultra- and nanofiltration membranes tested, only nanofiltration memebranes produced permeate whose quality was very close to that of the chemically treated raw water used as fresh water in most paper mills today and which should thus be well suited for reuse as shower water both in the wire and press section. It was also shown that a one stage nanofiltration process was more effective than processes in which micro- or ultrafiltration was used as pretreatment for nanofiltration. It was generally observed that acidic pH, high organic matter content, the presence of multivalent ions, hydrophobic membrane material and high membrane cutoff increased the fouling tendency of the membranes.