Strategies for sequential prediction of stationary time series

We present simple procedures for the prediction of a real valued sequence. The algorithms are based on a combinationof several simple predictors. We show that if the sequence is a realization of a bounded stationary and ergodic random process then the average of squared errors converges, almost surely, to that of the optimum, given by the Bayes predictor. We offer an analog result for the prediction of stationary gaussian processes.

Audit risk and rent extraction: Evidence from a randomized evaluation in Brazil

We report results from a randomized policy experiment designed to test whether increasedaudit risk deters rent extraction in local public procurement and service delivery in Brazil. Ourestimates suggest that temporarily increasing annual audit risk by about 20 percentage pointsreduced the proportion of irregular local procurement processes by about 17 percentage points.This reduction was driven entirely by irregularities involving mismanagement or corruption. Incontrast, we find no evidence that increased audit risk affected the quality of publicly providedpreventive and primary health care services -measured based on user satisfaction surveys- orcompliance with national regulations of the conditional cash transfer program "Bolsa Família".

A randomized concave programming method for choice network revenue management

Models incorporating more realistic models of customer behavior, as customers choosing froman offer set, have recently become popular in assortment optimization and revenue management.The dynamic program for these models is intractable and approximated by a deterministiclinear program called the CDLP which has an exponential number of columns. However, whenthe segment consideration sets overlap, the CDLP is difficult to solve. Column generationhas been proposed but finding an entering column has been shown to be NP-hard. In thispaper we propose a new approach called SDCP to solving CDLP based on segments and theirconsideration sets. SDCP is a relaxation of CDLP and hence forms a looser upper bound onthe dynamic program but coincides with CDLP for the case of non-overlapping segments. Ifthe number of elements in a consideration set for a segment is not very large (SDCP) can beapplied to any discrete-choice model of consumer behavior. We tighten the SDCP bound by(i) simulations, called the randomized concave programming (RCP) method, and (ii) by addingcuts to a recent compact formulation of the problem for a latent multinomial-choice model ofdemand (SBLP+). This latter approach turns out to be very effective, essentially obtainingCDLP value, and excellent revenue performance in simulations, even for overlapping segments.By formulating the problem as a separation problem, we give insight into why CDLP is easyfor the MNL with non-overlapping considerations sets and why generalizations of MNL posedifficulties. We perform numerical simulations to determine the revenue performance of all themethods on reference data sets in the literature.

Analgesic efficacy of intravenous perfusion of lidocaine, ketamine or a combination after laparotomy in a placebo-controlled, randomized, double-blind prospective study

In acute postoperative pain management intravenous lidocaine and/or ketamine have been advocated because of their morphine-sparing effect. The goal of this prospective, randomised, double-blind study was to assess morphine consumption with different regimens of intravenous infusion of lidocaine, ketamine or both during 48 hours following laparotomy. Patients were randomised into four groups. Group L, K, and KL received intravenous lidocaine, ketamine or a combination, respectively, before incision and during 48 hours postoperatively. The control group (C) received a similar volume of saline bolus and infusion. Postoperative analgesia included morphine delivered by a patient-controlled analgesia device. Primary outcome was the cumulative morphine consumption and pain, sedation scores, pressure algometry and side effects were our secondary outcomes. Cognition and psychomotor performance were also tested. Out of 57 eligible patients, 44 completed the study. Lidocaine reduced the cumulative morphine consumption compared with the control group (mean 0.456 mg.kg-1 +/- 0.244 (SD) versus 0.705 +/- 0.442, respectively, &#929; < 0.001). Pain scores during movement were statistically lower in all three treatment groups. Psychometric tests showed that the lidocaine group expressed more depressed feelings and sadness compared to the control group. Lidocaine administration had a morphine-sparing effect with a 36% reduction of morphine consumption while ketamine alone or combined with lidocaine did not. As a whole, our results suggest that intravenous lidocaine may offer advantages for postoperative analgesia. We propose lidocaine as a new alternative for pain control that needs to be studied further in future multicentric studies.

Markers of atherosclerosis and inflammation for prediction of coronary heart disease in older adults.

Although both inflammatory and atherosclerosis markers have been associated with coronary heart disease (CHD) risk, data directly comparing their predictive value are limited. The authors compared the value of 2 atherosclerosis markers (ankle-arm index (AAI) and aortic pulse wave velocity (aPWV)) and 3 inflammatory markers (C-reactive protein (CRP), interleukin-6 (IL-6), and tumor necrosis factor-alpha (TNF-alpha)) in predicting CHD events. Among 2,191 adults aged 70-79 years at baseline (1997-1998) from the Health, Aging, and Body Composition Study cohort, the authors examined adjudicated incident myocardial infarction or CHD death ("hard" events) and "hard" events plus hospitalization for angina or coronary revascularization (total CHD events). During 8 years of follow-up between 1997-1998 and June 2007, 351 participants developed total CHD events (197 "hard" events). IL-6 (highest quartile vs. lowest: hazard ratio = 1.82, 95% confidence interval: 1.33, 2.49; P-trend &lt; 0.001) and AAI (AAI &lt;/= 0.9 vs. AAI 1.01-1.30: hazard ratio = 1.57, 95% confidence interval: 1.14, 2.18) predicted CHD events above traditional risk factors and modestly improved global measures of predictive accuracy. CRP, TNF-alpha, and aPWV had weaker associations. IL-6 and AAI accurately reclassified 6.6% and 3.3% of participants, respectively (P's &lt;/= 0.05). Results were similar for "hard" CHD, with higher reclassification rates for AAI. IL-6 and AAI are associated with future CHD events beyond traditional risk factors and modestly improve risk prediction in older adults.

Efficacy of brief motivational intervention at army conscription to address alcohol use and related problems: a randomized controlled trial

Objectives: The aim of this study was to evaluate the efficacy of brief motivational intervention (BMI) in reducing alcohol use and related problems among binge drinkers randomly selected from a census of 20 year-old French speaking Swiss men and to test the hypothesis that BMI contributes to maintain low-risk drinking among non-bingers. Methods: Randomized controlled trial comparing the impact of BMI on weekly alcohol use, frequency of binge drinking and occurrence of alcohol-related problems. Setting: Army recruitment center. Participants: A random sample of 622 men were asked to participate, 178 either refused, or missed appointment, or had to follow military assessment procedures instead, resulting in 418 men randomized into BMI or control conditions, 88.7% completing the 6-month follow-up assessment. Intervention: A single face-to-face BMI session exploring alcohol use and related problems in order to stimulate behaviour change perspective in a non-judgmental, empathic manner based on the principles of motivational interviewing (MI). Main outcome measures: Weekly alcohol use, binge drinking frequency and the occurrence of 12 alcohol-related consequences. Results: Among binge drinkers, we observed a 20% change in drinking induced by BMI, with a reduction in weekly drinking of 1.5 drink in the BMI group, compared to an increase of 0.8 drink per week in the control group (incidence rate ratio 0.8, 95% confidence interval 0,66 to 0,98, p = 0.03). BMI did not influence the frequency of binge drinking and the occurrence of 12 possible alcohol-related consequences. However, BMI induced a reduction in the alcohol use of participants who, after drinking over the past 12 months, experienced alcohol-related consequences, i.e., hangover (-20%), missed a class (-53%), got behind at school (-54%), argued with friends (-38%), engaged in unplanned sex (-45%) or did not use protection when having sex (-64%). BMI did not reduce weekly drinking in those who experienced the six other problems screened. Among non-bingers, BMI did not contribute to maintain low-risk drinking. Conclusions: At army conscription, BMI reduced alcohol use in binge drinkers, particularly in those who recently experienced alcohol-related adverse consequences. No preventive effect of BMI was observed among non-bingers. BMI is an interesting preventive option in young binge drinkers, particularly in countries with mandatory army recruitment.

Leptin and smoking cessation: secondary analyses of a randomized controlled trial assessing physical activity as an aid for smoking cessation.

BACKGROUND: Smokers have a lower body weight compared to non-smokers. Smoking cessation is associated with weight gain in most cases. A hormonal mechanism of action might be implicated in weight variations related to smoking, and leptin might be implicated. We made secondary analyses of an RCT, with a hypothesis-free exploratory approach to study the dynamic of leptin following smoking cessation. METHODS: We measured serum leptin levels among 271 sedentary smokers willing to quit who participated in a randomized controlled trial assessing a 9-week moderate-intensity physical activity intervention as an aid for smoking cessation. We adjusted leptin for body fat levels. We performed linear regressions to test for an association between leptin levels and the study group over time. RESULTS: One year after smoking cessation, the mean serum leptin change was +3.23 mg/l (SD 4.89) in the control group and +1.25 mg/l (SD 4.86) in the intervention group (p of the difference&#8201;<&#8201;0.05). When adjusted for body fat levels, leptin was higher in the control group than in the intervention group (p of the difference&#8201;<&#8201;0.01). The mean weight gain was +2.91 (SD 6.66) Kg in the intervention and +3.33 (SD 4.47) Kg in the control groups, respectively (p not significant). CONCLUSIONS: Serum leptin levels significantly increased after smoking cessation, in spite of substantial weight gain. The leptin dynamic might be different in chronic tobacco users who quit smoking, and physical activity might impact the dynamic of leptin in such a situation. CLINICAL TRIAL REGISTRATION NUMBER: NCT00521391.

Robust response transformations based on optimal prediction

Nonlinear regression problems can often be reduced to linearity by transforming the response variable (e.g., using the Box-Cox family of transformations). The classic estimates of the parameter defining the transformation as well as of the regression coefficients are based on the maximum likelihood criterion, assuming homoscedastic normal errors for the transformed response. These estimates are nonrobust in the presence of outliers and can be inconsistent when the errors are nonnormal or heteroscedastic. This article proposes new robust estimates that are consistent and asymptotically normal for any unimodal and homoscedastic error distribution. For this purpose, a robust version of conditional expectation is introduced for which the prediction mean squared error is replaced with an M scale. This concept is then used to develop a nonparametric criterion to estimate the transformation parameter as well as the regression coefficients. A finite sample estimate of this criterion based on a robust version of smearing is also proposed. Monte Carlo experiments show that the new estimates compare favorably with respect to the available competitors.

Accounting information and the prediction of farm viability

Until recently farm management made little use of accounting and agriculture has been largely excluded from the scope of accounting standards. This article examines the current use of accounting in agriculture and points theneed to establish accounting standards for agriculture. Empirical evidence shows that accounting can make a significant contribution to agricultural management and farm viability and could also be important for other agents involved in agricultural decision making. Existing literature on failureprediction models and farm viability prediction studies provide the starting point for our research, in which two dichotomous logit models were applied to subsamples of viable and unviable farms in Catalonia, Spain. The firstmodel considered only non-financial variables, while the other also considered financial ones. When accounting variables were added to the model, a significant reduction in deviance was observed.

Improving generalized regression analysis for the spatial prediction of forest communities

Aim This study used data from temperate forest communities to assess: (1) five different stepwise selection methods with generalized additive models, (2) the effect of weighting absences to ensure a prevalence of 0.5, (3) the effect of limiting absences beyond the environmental envelope defined by presences, (4) four different methods for incorporating spatial autocorrelation, and (5) the effect of integrating an interaction factor defined by a regression tree on the residuals of an initial environmental model. Location State of Vaud, western Switzerland. Methods Generalized additive models (GAMs) were fitted using the grasp package (generalized regression analysis and spatial predictions, http://www.cscf.ch/grasp). Results Model selection based on cross-validation appeared to be the best compromise between model stability and performance (parsimony) among the five methods tested. Weighting absences returned models that perform better than models fitted with the original sample prevalence. This appeared to be mainly due to the impact of very low prevalence values on evaluation statistics. Removing zeroes beyond the range of presences on main environmental gradients changed the set of selected predictors, and potentially their response curve shape. Moreover, removing zeroes slightly improved model performance and stability when compared with the baseline model on the same data set. Incorporating a spatial trend predictor improved model performance and stability significantly. Even better models were obtained when including local spatial autocorrelation. A novel approach to include interactions proved to be an efficient way to account for interactions between all predictors at once. Main conclusions Models and spatial predictions of 18 forest communities were significantly improved by using either: (1) cross-validation as a model selection method, (2) weighted absences, (3) limited absences, (4) predictors accounting for spatial autocorrelation, or (5) a factor variable accounting for interactions between all predictors. The final choice of model strategy should depend on the nature of the available data and the specific study aims. Statistical evaluation is useful in searching for the best modelling practice. However, one should not neglect to consider the shapes and interpretability of response curves, as well as the resulting spatial predictions in the final assessment.

Laparoscopic gastric banding: a prospective, randomized study comparing the Lapband and the SAGB: early results.

OBJECTIVE: The objective of this study was to evaluate the results of laparoscopic gastric banding using 2 different bands (the Lapband [Bioenterics, Carpinteria, CA] and the SAGB [Swedish Adjustable Gastric Band; Obtech Medical, 6310 Zug, Switzerland]) in terms of weight loss and correction of comorbidities, short-and long-term complications, and improvement of quality of life in morbidly obese patients SUMMARY BACKGROUND DATA: During the past 10 years, gastric banding has become 1 of the most common bariatric procedures, at least in Europe and Australia. Weight loss can be excellent, but it is not sufficient in a significant proportion of patients, and a number of long-term complications can develop. We hypothesized that the type of band could be of importance in the outcome. METHODS: One hundred eighty morbidly obese patients were randomly assigned to receive the Lapband or the SAGB. All the procedures were performed by the same surgeon. The primary end point was weight loss, and secondary end points were correction of comorbidities, early- and long-term complications, importance of food restriction, and improvement of quality of life. RESULTS: Initial weight loss was faster in the Lapband group, but weight loss was eventually identical in the 2 groups. There was a trend toward more early band-related complications and more band infections with the SAGB, but the study had limited power in that respect. Correction of comorbidities, food restriction, long-term complications, and improvement of quality of life were identical. Only 55% to 60% of the patients achieved an excess weight loss of at least 50% in both groups. There was no difference in the incidence of long-term complications. CONCLUSIONS: Gastric banding can be performed safely with the Lapband or the SAGB with similar short- and midterm results with respect to weight loss and morbidity. Only 50% to 60% of the patients will achieve sufficient weight loss, and close to 10% at least will develop severe long-term complications.

Promoting rule compliance in daily-life: Evidence from a randomized field experiment in the public libraries of Barcelona

We study how to promote compliance with rules in everyday situations. Having access to unique data on the universe of users of all public libraries inBarcelona, we test the effect of sending email messages with dierent contents.We find that users return their items earlier if asked to do so in a simple email.Emails reminding users of the penalties associated with late returns are more effective than emails with only a generic reminder. We find differential treatmenteffects by user types. The characteristics we analyze are previous compliance,gender, age, and nationality.

Influence of CYP2D6 activity on pre-emptive analgesia by the N-methyl-D-aspartate antagonist dextromethorphan in a randomized controlled trial of acute pain.

BACKGROUND: There is some evidence that dextromethorphan (DM) is effective as a pre-emptive analgesic agent.  DM is mainly metabolized to dextrorphan (DOR) by CYP2D6 whose activity can be inhibited by pharmacologic intervention. OBJECTIVES: To investigate the efficacy of DM as a pre-emptive analgesic agent and describe the population pharmacokinetics in the presence of normal and poor CYP2D6 metabolism in acute post-operative pain. STUDY DESIGN: Double blind, randomized, placebo-controlled trial SETTING: Post-surgical analgesic consumption after knee ligament surgery, a setting of acute pain. METHODS: Forty patients were randomized to a single oral dose of 50 mg quinidine or placebo, administered 12 hours before 50 mg DM. Patients were genotyped for the major CYP2D6 and ABCB1 variants and phenotyped for CYP2D6 using urine DM/DOR metabolic ratios and blood samples for population pharmacokinetic modeling. RESULTS: Quinidine was effective in inhibiting CYP2D6 activity, with 2-fold reduction of DM to DOR biotransformation clearance, prolonged DM half-life, and increased DM systemic availability. Patients in the quinidine group required significantly less often NSAIDs than patients in the placebo group (35.3% vs. 75.0%, P = 0.022). The odds ratio for NSAID consumption in the placebo vs. quinidine group was 5.5 (95% confidence interval (CI) 1.3 - 22.7) at 48 hours after surgery. LIMITATIONS: While this study shows an impact of DM on pre-emptive analgesia and is mechanistically interesting, the findings need to be confirmed in larger trials. CONCLUSION: CYP2D6 inhibition by quinidine influenced the pre-emptive analgesic effectiveness of DM confirming that CYP2D6 phenotypic switch increases the neuromodulatory effect of oral dextromethorphan.