The Equiprobability Bias from a Mathematical and Psychological Perspective

The equiprobability bias (EB) is a tendency to believe that every process in which randomness is involved corresponds to a fair distribution, with equal probabilities for any possible outcome. The EB is known to affect both children and adults, and to increase with probability education. Because it results in probability errors resistant to pedagogical interventions, it has been described as a deep misconception about randomness: the erroneous belief that randomness implies uniformity. In the present paper, we show that the EB is actually not the result of a conceptual error about the definition of randomness. On the contrary, the mathematical theory of randomness does imply uniformity. However, the EB is still a bias, because people tend to assume uniformity even in the case of events that are not random. The pervasiveness of the EB reveals a paradox: The combination of random processes is not necessarily random. The link between the EB and this paradox is discussed, and suggestions are made regarding educational design to overcome difficulties encountered by students as a consequence of the EB.

Outperformance in exchange-traded fund pricing deviations: Generalized control of data snooping bias

An investigation into exchange-traded fund (ETF) outperforrnance during the period 2008-2012 is undertaken utilizing a data set of 288 U.S. traded securities. ETFs are tested for net asset value (NAV) premium, underlying index and market benchmark outperformance, with Sharpe, Treynor, and Sortino ratios employed as risk-adjusted performance measures. A key contribution is the application of an innovative generalized stepdown procedure in controlling for data snooping bias. We find that a large proportion of optimized replication and debt asset class ETFs display risk-adjusted premiums with energy and precious metals focused funds outperforming the S&P 500 market benchmark. 

Practical Spectrum Aggregation for Secondary Networks with Imperfect Sensing

We investigate a collision-sensitive secondary network that intends to opportunistically aggregate and utilize spectrum of a primary network to achieve higher data rates. In opportunistic spectrum access with imperfect sensing of idle primary spectrum, secondary transmission can collide with primary transmission. When the secondary network aggregates more channels in the presence of the imperfect sensing, collisions could occur more often, limiting the performance obtained by spectrum aggregation. In this context, we aim to address a fundamental query, that is, how much spectrum aggregation is worthy with imperfect sensing. For collision occurrence, we focus on two different types of collision: one is imposed by asynchronous transmission; and the other by imperfect spectrum sensing. The collision probability expression has been derived in closed-form with various secondary network parameters: primary traffic load, secondary user transmission parameters, spectrum sensing errors, and the number of aggregated sub-channels. In addition, the impact of spectrum aggregation on data rate is analysed under the constraint of collision probability. Then, we solve an optimal spectrum aggregation problem and propose the dynamic spectrum aggregation approach to increase the data rate subject to practical collision constraints. Our simulation results show clearly that the proposed approach outperforms the benchmark that passively aggregates sub-channels with lack of collision awareness.

Failure of a numerical quality assessment scale to identify potential risk of bias in a systematic review: a comparison study

BACKGROUND: Assessing methodological quality of primary studies is an essential component of systematic reviews. Following a systematic review which used a domain based system [United States Preventative Services Task Force (USPSTF)] to assess methodological quality, a commonly used numerical rating scale (Downs and Black) was also used to evaluate the included studies and comparisons were made between quality ratings assigned using the two different methods. Both tools were used to assess the 20 randomized and quasi-randomized controlled trials examining an exercise intervention for chronic musculoskeletal pain which were included in the review. Inter-rater reliability and levels of agreement were determined using intraclass correlation coefficients (ICC). Influence of quality on pooled effect size was examined by calculating the between group standardized mean difference (SMD).

RESULTS: Inter-rater reliability indicated at least substantial levels of agreement for the USPSTF system (ICC 0.85; 95% CI 0.66, 0.94) and Downs and Black scale (ICC 0.94; 95% CI 0.84, 0.97). Overall level of agreement between tools (ICC 0.80; 95% CI 0.57, 0.92) was also good. However, the USPSTF system identified a number of studies (n = 3/20) as "poor" due to potential risks of bias. Analysis revealed substantially greater pooled effect sizes in these studies (SMD -2.51; 95% CI -4.21, -0.82) compared to those rated as "fair" (SMD -0.45; 95% CI -0.65, -0.25) or "good" (SMD -0.38; 95% CI -0.69, -0.08).

CONCLUSIONS: In this example, use of a numerical rating scale failed to identify studies at increased risk of bias, and could have potentially led to imprecise estimates of treatment effect. Although based on a small number of included studies within an existing systematic review, we found the domain based system provided a more structured framework by which qualitative decisions concerning overall quality could be made, and was useful for detecting potential sources of bias in the available evidence.

Data-Augmentation for Reducing Dataset Bias in Person Re-identification

In this paper we explore ways to address the issue of dataset bias in person re-identification by using data augmentation to increase the variability of the available datasets, and we introduce a novel data augmentation method for re-identification based on changing the image background. We show that use of data augmentation can improve the cross-dataset generalisation of convolutional network based re-identification systems, and that changing the image background yields further improvements.

Factors influencing success of clinical genome sequencing across a broad spectrum of disorders

To assess factors influencing the success of whole-genome sequencing for mainstream clinical diagnosis, we sequenced 217 individuals from 156 independent cases or families across a broad spectrum of disorders in whom previous screening had identified no pathogenic variants. We quantified the number of candidate variants identified using different strategies for variant calling, filtering, annotation and prioritization. We found that jointly calling variants across samples, filtering against both local and external databases, deploying multiple annotation tools and using familial transmission above biological plausibility contributed to accuracy. Overall, we identified disease-causing variants in 21% of cases, with the proportion increasing to 34% (23/68) for mendelian disorders and 57% (8/14) in family trios. We also discovered 32 potentially clinically actionable variants in 18 genes unrelated to the referral disorder, although only 4 were ultimately considered reportable. Our results demonstrate the value of genome sequencing for routine clinical diagnosis but also highlight many outstanding challenges.