The multiple filter test for change point detection in time series

A framework for the detection of change points in the expectation in sequences of random variables is presented. Specifically, we investigate time series with general distributional assumptions that may show an unknown number of change points in the expectation occurring on multiple time scales and that may also contain change points in other parameters. To that end we propose a multiple filter test (MFT) that tests the null hypothesis of constant expectation and, in case of rejection of the null hypothesis, an algorithm that estimates the change points.The MFT has three important benefits. First, it allows for general distributional assumptions in the underlying model, assuming piecewise sequences of i.i.d. random variables, where also relaxations with regard to identical distribution or independence are possible. Second, it uses a MOSUM type statistic and an asymptotic setting in which the MOSUM process converges weakly to a functional of a Brownian motion which is then used to simulate the rejection threshold of the statistical test. This approach enables a simultaneous application of multiple MOSUM processes which improves the detection of change points that occur on different time scales. Third, we also show that the method is practically robust against changes in other distributional parameters such as the variance or higher order moments which might occur with or even without a change in expectation. A function implementing the described test and change point estimation is available in the R package MFT.     

An Experimental Analysis of Modifications to the Centralized Milk Quota Exchange System in Quebec

Using experimental economics, this paper tests the potential impacts of modifying the centralized quota exchange system in Quebec with the intent of decreasing the quota price while minimizing negative impacts on auction effectiveness. Two separate treatments are applied to a uniform price auction similar to that employed in Quebec. The first treatment is an exclusion (5% or 15%) of the highest buyer bids and seller offers. The second is a tax (2% or 10%) on all units offered for sale that remain unsold. Various combinations of the two treatments are also tested. The results suggest that exclusion of the highest bids and offers can decrease the price of the quota and that a 15% exclusion rate is more effective than a 5% rate. The tax alone has little impact on quota price. The combination of the two treatments generates a more marked reduction in both the number of exchanges and the price of the quota than when the tax or the exclusion is applied individually. However, the combination of treatments results in a greater loss of economic efficiency. In all cases, relatively small market price reductions are realized at the expense of substantial losses in economic efficiency. La présente étude teste de manière expérimentale la capacité d'une modification au système centralisé de vente du quota à faire diminuer le prix du quota laitier au Québec tout en minimisant les impacts négatifs des changements sur l'efficacité de l'enchère. Cette modification consiste à appliquer deux traitements sur l'enchère de prix uniforme où s'échange le quota. Le premier traitement consiste en une exclusion (5% ou 15%) des mises les plus élevées des acheteurs et des vendeurs. Le second traitement est une taxe (2% ou 10%) appliquée aux unités que les vendeurs mettent en marché et ne réussissent pas à vendre. Différentes combinaisons de ces deux traitements sont également testées. Les données générées permettent de conclure que le mécanisme d'exclusion des mises les plus élevées permet de faire diminuer le prix du quota, l'exclusion de 15%étant plus efficace que celle de 5%. Pour sa part, la taxe seule a peu d'impact sur le prix du quota tandis que la combinaison des deux traitements entraîne une diminution du nombre de transactions et du prix du quota plus marquée que lorsque les traitements de taxe et d'exclusion sont appliqués individuellement. Cela a comme corollaire d'entraîner une perte d'efficacitééconomique plus importante. Dans tous les cas, des baisses de prix de marché relativement modestes sont réalisées au coût d'importantes pertes d'efficacitééconomique.     

Coerced Confessions: Self-Policing in the Shadow of the Regulator

As part of a recent trend toward more cooperative relationsbetween regulators and industry, novel government programs areencouraging firms to monitor their own regulatory complianceand voluntarily report their own violations. In this study,we examine how regulatory enforcement activities influence organizations'decisions to self-police. We created a comprehensive data setfor the "Audit Policy," a United States Environmental ProtectionAgency program that encourages companies to self-disclose violationsof environmental laws and regulations in exchange for reducedsanctions. We find that facilities are more likely to self-discloseif they were recently subjected to one of several differentenforcement measures and if they were provided with immunityfrom prosecution for self-disclosed violations.     

Repeated payment delinquency among young adults in the United States

There is concern in the United States about young adults falling behind financially due to the increased use of student loans and low wages. This study investigates payment delinquency as a measure of financial distress to better understand how young adults might be struggling. Personality traits are incorporated into the model to determine the extent to which behavioural factors are correlated with financial behaviors and if they predict a habit trend of payment delinquency. The 1997 National Longitudinal Survey of Youth, a nationally representative longitudinal data set, is used in the study. A random effects probit model and a dynamic random effects probit model are used to examine late bill pay (harassed by bill collectors) and late rent or mortgage payments (more than 60 days late) over a period of 8 years (2007–2015). Results from the analysis indicate that payment delinquency in a previous period increases the likelihood of payment delinquency by 10 percentage points in a subsequent period. Conscientiousness decreases the likelihood by 2.1 percentage points, while neuroticism increases the likelihood by 1.6 percentage points.     

Living conditions of mine workers from eight mines in South Africa

Poor living conditions are a consequence of the history of the South African mining industry (SAMI), despite legislation having been implemented to attempt to address this challenge. This paper describes the living conditions of mine workers from eight mines in South Africa in 2014, and assesses changes made over the previous decade. Qualitative and quantitative data were collected at three platinum, two gold, one coal, one diamond, and one manganese mine in 53 interviews with mine management, 11 interviews with labour representatives, 14 focus groups with mine workers, and 875 questionnaires completed by mine workers. The use of single-sex hostels and hostel room occupancy rates has reduced, while the use of living-out allowances (LOAs) has increased. Problems included the high proportions of informal accommodation; a lack of access to water, sanitation, and electricity; and poor roads. While improvements to the living conditions in the SAMI are evident, challenges still remain.     

Social preferences and voting: An exploration using a novel preference revealing mechanism

Public referenda are frequently used to determine the provision of public goods. As public programs have distributional consequences, a compelling question is what impact, if any, do social preferences have on voting behavior. This paper explores this issue using laboratory experiments wherein voting outcomes lead to a known distribution of net benefits across participants. Preferences are elicited using a novel Random Price Voting Mechanism (RPVM), which is more efficient in eliciting preferences than a dichotomous choice referendum but gives consistent results. Results suggest that social preferences, in particular a social efficiency motive, lead to economically meaningful deviations from selfish voting choices and increase the likelihood that welfare-enhancing programs are implemented.     

Can Stigma Explain Large Property Value Losses? The Psychology and Economics of Superfund

This research documents the long term impacts of delayed cleanup on property values in communities neighboring prominent Superfund sites. The research examines the sale prices of nearly 34,000 homes near sites in three metropolitan areas for up to a 30-year period. To our knowledge, no other property value studies have examined sites in multiple areas with large property value losses over the length of time used here. The results are both surprising and inconsistent with most prior work. The principal result is that, when cleanup is delayed for 10, 15, and even up to 20 years, the discounted present value of the cleanup is mostly lost. A possible explanation for these property value losses is that the sites are stigmatized and the homes in the surrounding communities are shunned. The results suggest that expedited cleanup and minimizing the number of stigmatizing events would reduce these losses. This research was supported by the USEPA under cooperative agreement CR 824393-01-0. The authors wish to thank Alan Carlin for his patience and support along with Shelby Gerking, Kip Viscusi and three anonymous reviewers for their thoughtful comments. We also would like to thank Christian Coerds, Graham Crawford, Rachel Deming, Karen Grace-Martin, Brian Hurd, Joe LaVerde, Eleanor Smith, and Matt Todaro for their support on this project and the participants of the Risk Perception, Valuation and Policy conference at the University of Central Florida and the AERE Workshop in Estes Park, Colorado, for their helpful feedback.     

Voluntary Funding for Generic Advertising Using a Provision Point Mechanism: An Experimental Analysis of Option Assurance

The mandatory nature of check-off programs remains a contentious policy issue. A provision point mechanism with a money-back guarantee offers an attractive, alternative voluntary funding approach, yet faces the problem of how to maintain future administrative capacity should contributions fall short of the funding threshold in the present period. To address this concern, a novel two-threshold, provision point mechanism is tested that sets a high threshold to fund marketing and a low threshold, if necessary, to secure administrative capacity for future advertising. Experimental results demonstrate that providing such "option assurance" sustains high overall contributions and, in some settings, can increase producer surplus.     

Context Effects in a Negatively Framed Social Dilemma Experiment

A number of studies of the positively-framed voluntary contribution mechanism (VCM) public goods game have examined the role of context on participant behavior. Relatively little attention, however, has been given to the effects of context in negatively-framed VCM games even though this setting is relevant to a wide array of real world challenges, such as common pool resource use and pollution externalities. This study uses a carefully-controlled laboratory experiment to investigate the degree to which the context in which decisions are made influences decision making in a negatively-framed VCM laboratory experiment. The context treatments that we evaluate vary communication, voting, and the status quo of the initial endowment. Results indicate that providing groups the opportunity to communicate and vote significantly reduces choices that impose external costs. Importantly, the pro-social effects of communication and voting are strongest when the status quo endowment is the private account, which generates costs on other participants. This result suggests that the effect of the status quo endowment is a function of whether the social dilemma is framed positively or negatively when communication between participants is allowed.     

Measuring Stigma: The Behavioral Implications of Disgust

Stigmatization of products and technologies can lead to large monetary losses even when there are no associative risks. This paper reports on experiments that provide insight into the behavioral responses of disgust from an economic perspective. We use a dead sterilized cockroach to ‘contaminate’ drinking water and generate willingness-to-pay (WTP) and willingness-to-accept (WTA) measures of participants’ reactions. These results are contrary to previous results from research not involving financial incentives, as most participants’ WTP and WTA values are near zero for drinking cockroach contaminated water. Additionally, filtration of cockroach water leads participants to become significantly more likely to request compensation compared to spring water, but it does not result in requesting significantly more money to drink it. Finally, WTP and WTA differences can be explained by participants’ decision on whether or not to request compensation and not by the amount of compensation.