Upward Bias Research Articles

Performance of compressed sensing for fluorine-19 magnetic resonance imaging at low signal-to-noise ratio conditions.

To examine the performance of compressed sensing (CS) in reconstructing low signal-to-noise ratio (SNR) 19 F MR signals that are close to the detection threshold and originate from small signal sources with no a priori known location. Regularization strength was adjusted automatically based on noise level. As performance metrics, root-mean-square deviations, true positive rates (TPRs), and false discovery rates were computed. CS and conventional reconstructions were compared at equal measurement time and evaluated in relation to high-SNR reference data. 19 F MR data were generated from a purpose-built phantom and benchmarked against simulations, as well as from the experimental autoimmune encephalomyelitis mouse model. We quantified the signal intensity bias and introduced an intensity calibration for in vivo data using high-SNR ex vivo data. Low-SNR 19 F MR data could be reliably reconstructed. Detection sensitivity was consistently improved and data fidelity was preserved for undersampling and averaging factors of α = 2 or = 3. Higher α led to signal blurring in the mouse model. The improved TPRs at α = 3 were comparable to a 2.5-fold increase in measurement time. Whereas CS resulted in a downward bias of the 19 F MR signal, Fourier reconstructions resulted in an unexpected upward bias of similar magnitude. The calibration corrected signal-intensity deviations for all reconstructions. CS is advantageous whenever image features are close to the detection threshold. It is a powerful tool, even for low-SNR data with sparsely distributed 19 F signals, to improve spatial and temporal resolution in 19 F MR applications.

Using a Cost of Capital Where One Really Shouldn't

Myers (1974---‘M74’) derives the Cost of Capital (CC), including the corporate-tax subsidy on borrowing, for a one-period investment project. From Miles and Ezzell (1980), this CC remains valid in multiperiod projects provided that, during the project's life, debt is continuously re-aligned with ex post GPVs. In contrast, when the debt schedule is static --- i.e. set to stabilise leverage ex ante only --- no CC works outside the Modigliani and Miller (1963---‘MM63’) perpetuities scenario; one generally needs a recursive Adjusted NPV (rANPV). Real-life leverage being neither fully static nor fully dynamic, two questions arise: (i) do M74's CC and rANPV produce very different valuations in a static/multiperiod problem?, and (ii) is it useful to heuristically adopt a non-linear interpolation between the M74 and MM63 polar CCs? We find, first, that M74's CC is is (mildly) conservative but vastly less biased and more precise than MM63's. Second, two of the heuristic compromise CCs are marginally more correct than even M74, but they remain biased upwards. Remembering that M74 also applies when debt is GPV-indexed ex post, M74 seems the safer procedure. Highlights: – With a ‘static’ debt schedule set to stabilise leverage ex ante, a finite-life project’s risk cannot be constant, so no Cost of Capital (CC) can work perfectly. – When expressed as fractions of gross present value, minor-looking CC-based valuation errors largely reflect the low PV of the true tax shields rather than a precise assessment. – The Myers (1974) one-period cost of capital (CC) is mildly conservative and works very well for multiperiod NPV-ing even when debt is static. – The Miller-Modigliani CC for perpetuities, when applied in finite-life problems, is vastly overoptimistic and imprecise, especially when the CC takes into account growth. – We also propose four heuristic procedures that non-linearly interpolate between the above two CCs. The two that take into account the changing risk do succeed in beating Myers’ CC, but the gains are small and the bias is upward rather than conservative.

Classification of brain tumor isocitrate dehydrogenase status using MRI and deep learning.

Isocitrate dehydrogenase (IDH) mutation status is an important marker in glioma diagnosis and therapy. We propose an automated pipeline for noninvasively predicting IDH status using deep learning and T2-weighted (T2w) magnetic resonance (MR) images with minimal preprocessing (N4 bias correction and normalization to zero mean and unit variance). T2w MR images and genomic data were obtained from The Cancer Imaging Archive dataset for 260 subjects (120 high-grade and 140 low-grade gliomas). A fully automated two-dimensional densely connected model was trained to classify IDH mutation status on 208 subjects and tested on another held-out set of 52 subjects using fivefold cross validation. Data leakage was avoided by ensuring subject separation during the slice-wise randomization. Mean classification accuracy of 90.5% was achieved for each axial slice in predicting the three classes of no tumor, IDH mutated, and IDH wild type. Test accuracy of 83.8% was achieved in predicting IDH mutation status for individual subjects on the test dataset of 52 subjects. We demonstrate a deep learning method to predict IDH mutation status using T2w MRI alone. Radiologic imaging studies using deep learning methods must address data leakage (subject duplication) in the randomization process to avoid upward bias in the reported classification accuracy.

The effects of subordinates’ use of upward influence tactics on their supervisors’ job performance evaluations in Saudi Arabia: the significance of loyalty

Most research examining the effects of subordinates’ use of upward influence tactics on supervisor job performance evaluations has been conducted in Western countries. In these contexts, it has been suggested that upward influence tactics bias supervisor ratings because they affect the quality of the relationship between supervisors and subordinates. This has primarily been explained in terms of supervisor liking. We suggest instead, that the particular cultural context in Saudi Arabia emphasises loyalty as the primary indicator of relationship quality. Based on data we obtained from 389 matched supervisor–subordinate dyads in Saudi Arabia, we found that five upward influence tactics; rational persuasion, ingratiation, self-promotion, coalition and upward appeal had indirect effects on supervisor job performance ratings through subordinate loyalty. Our findings suggest cultural contexts can emphasise differing aspects of relational quality between supervisors and subordinates that potentially explain bias in supervisor job performance evaluations.

Unbiased Estimation of Linkage Disequilibrium from Unphased Data.

Linkage disequilibrium (LD) is used to infer evolutionary history, to identify genomic regions under selection, and to dissect the relationship between genotype and phenotype. In each case, we require accurate estimates of LD statistics from sequencing data. Unphased data present a challenge because multilocus haplotypes cannot be inferred exactly. Widely used estimators for the common statistics r2 and D2 exhibit large and variable upward biases that complicate interpretation and comparison across cohorts. Here, we show how to find unbiased estimators for a wide range of two-locus statistics, including D2, for both single and multiple randomly mating populations. These unbiased statistics are particularly well suited to estimate effective population sizes from unlinked loci in small populations. We develop a simple inference pipeline and use it to refine estimates of recent effective population sizes of the threatened Channel Island Fox populations.

Improved estimators of correlation and R2 for skewed hydrologic data

ABSTRACTThe coefficient of determination R2 and Pearson correlation coefficient ρ = R are standard metrics in hydrology for the evaluation of the goodness of fit between model simulations and observations, and as measures of the degree of dependence of one variable upon another. We show that the standard product moment estimator of ρ, termed r, while well-behaved for bivariate normal data, is upward biased and highly variable for bivariate non-normal data. We introduce three alternative estimators of ρ which are nearly unbiased and exhibit much less variability than r for non-normal data. We also document remarkable upward bias and tremendous increases in variability associated with r using both synthetic data and daily streamflow simulations from 905 calibrated rainfall–runoff models. We show that estimators of ρ = R accounting for skewness are needed for daily streamflow series because they exhibit high variability and skewness compared to, for example, monthly/annual series, where r should perform well.

Proximity Bias Following Affective Metaphors in Patients With Depression-Psychoanalytic Considerations.

Background: Many languages use spatial metaphors to describe affective states such as an upward bias to denote positive mood, a downward bias to denote negative mood, a body proximity bias to denote personal relatedness concern, and a right-left bias to denote negative or positive valence. These biases might be related to experiential traces related to these affective states. If this is the case, depressed subjects would show either a downward spatial bias, a body proximity bias, or a right-left shift in attention. We evaluated the occurrence of such biases in subjects with depression compared to healthy controls.Methods: Subjects: 10 subjects with depression (5F:5M; age = 47.2 ± 15.2) and 10 healthy controls (5F:5M; age = 45.8 ± 14.5). Experimental task: line bisection task. Lines were presented in three spatial orientations [vertical (up-down), horizontal (right-left), radial (proximal-distal)] and were either blank, composed with words (negative/positive/neutral), or with smileys (negative/positive/neutral). There were 21 line types, and each was presented eight times, reaching a total of 168 lines.Results: Compared with healthy controls, subjects with depression bisected radial lines significantly closer to their body. There were no significant differences for either horizontal or vertical lines.Conclusion: The proximity spatial bias observed in subjects with depression suggests that depression might activate neural spatial networks. We argue that these networks could be dynamically activated through narcissistic mechanisms as implied in “Mourning and Melancholia” where Freud postulates a narcissistic mediated bias in depression according to which the depressed subjects withdraw from the outside world.

Energy Performance Certificates and Its Capitalization in Housing Values in Sweden

The impact on energy performance certificates on housing prices has been investigated extensively in recent years. However, the results of these investigations are mixed. We add to the literature by more specifically controlling for potential biases, by employing a combination of alternative approaches to estimate the causal relationship between house prices and energy performance certificates. We use a traditional hedonic modeling approach, but we additionally employ propensity score methods to be able to compare treated houses with a control group. We also investigate the impact of the outliers, spatial dependency, and parameter heterogeneity of our estimates. Moreover, we use the quantile regression technique to test the hypothesis that the capitalization effect varies across the price distribution. Our results, analyzing more than 100,000 observations, indicate there is an upward bias if one is not controlling for outlier and selection bias. Regardless of the propensity score method approach, the results are lower than a model (around 3 percent capitalization, compared to 6 percent). However, our results do not support that the impact of energy performance certificates varies in the price distribution. Consequently, the certificates are not differently capitalized in the high-end housing price segment. Finally, our results support the hypothesis that the energy performance certificate should be more capitalized into house prices in the northern and colder parts of Sweden than in the southern regions.

Observation of Wind-Loading Influences in Nonconcentric Radial Root Growth in Two Maple Species

In 2010 and 2016, Acer saccharinum and Acer rubrum roots were harvested and processed into transverse serial sections to observe crosssectional radial growth patterning in response to wind. Trees on the edge of a plantation and from interior positions were selected, and windward/leeward roots were targeted for a comparative assessment. While some observations were suggestive of a response to wind exposure, they were not definitive. Particularly in the windward versus leeward comparison within either edge or interior ground in terms of root size or radial growth pattern, there were no differences observed. In general, the loss of observed upward radial growth bias very closely coincided with the ending of the Zone of Rapid Taper in the architecture of the root plate.

The double EMG distribution and trade elasticities

AbstractUsing Brazilian export data that, unlike many trade data sets, have a full record of small export sales, this paper reconsiders trade elasticities and the welfare gains from trade. Using the Brazilian data, this paper provides novel evidence on the properties of the distributions of log‐export sales and shows that the double exponentially modified Gaussian (EMG) distribution parsimoniously captures these properties. Using the double EMG distribution in a standard monopolistic competition model of trade, this paper demonstrates that data truncation, which is prevalent in many data sets, leads to an upward bias in measuring the partial elasticity of trade with respect to variable trade costs. This bias subsequently leads to the underestimation of the gains from trade by 1% to 9% depending on the extent of data truncation, a range that is commensurate with typical economic growth and large booms.

The influence of traumatic brain injury on the allocation of vertical spatial attention

ABSTRACTObjective: Research on impairments of spatial attention has primarily investigated hemispatial neglect in brain-lesioned patients, revealing decrements in the allocation of attention to right versus left egocentric or allocentric hemispace. Whereas head trauma might injure those parts of the brain that allocate vertical attention, little is known about the influence of trauma on the allocation of visuospatial attention in vertical space. Thus, the goal of this study was to learn if chronic moderate-to-severe traumatic brain injury (m/sTBI) alters the allocation of vertical visuospatial attention as assessed by the Attention Network Task (ANT). The ANT assesses the influence of Posner-type spatial cues and distractors using an Eriksen flanker task.Methods: 12 chronic m/sTBI patients and 12 demographically-matched neurologically-healthy controls (HCs) completed a version of the ANT wherein trials were coded for cue and target locations above and below central visual fixation. Trial-wise reaction times (RT) and accuracy were subjected to mixed-model ANOVAs and planned contrasts.Results: These data were subject to secondary analyses, which revealed that across groups, median RTs were significantly faster when targets occurred above than below the central visual fixation (p < .01); however, only HCs’ error rates differed as a function of target altitude. Unlike controls, m/sTBI survivors did not exhibit the anticipated upward error-rate attentional bias.Conclusions: As alteration of spatial attention can be a major cause of disability, present findings suggest that m/sTBI survivors exhibit this loss of normal upward attentional bias. Future studies are need to learn if these patients might benefit from treatment.

The heterogenous impact of taxation on FDI: A note on Djankov et al. (2010)

Summary:Taxes in the host country matter a lot for FDI inflows, but only for non-OECD countries. Taking advantage of the rich dataset constructed by Djankov et al. (2010), we show, for example, that raising the first-year effective corporate income tax rate by 10 pp reduces FDI inflows by 3.4 to 1.9 pp in non-OECD countries; and the effect is null for OECD countries. Not taking this heterogeneity into account upward (downward) biases the estimated impact of the corporate tax on FDI inflows for OECD (non-OECD) countries.

Credit constraints and productivity of SMEs: Evidence from Canada

To what extent firms are constrained by external credit is usually unobserved in commonly used firm-level data. We use a survey of financing among Canadian small and medium-sized enterprises to measure the likelihood of a firm being constrained by credit. We find that firm size, current debt-to-asset ratio and cash flow are robust indicators of being financially constrained, while long-term debt to asset ratio is not a significant indicator of credit constraints. We then estimate the firm-level total factor productivity, taking into account the measured credit constraints. Omitting credit constraints leads to an upward bias of productivity estimates, by 4 percent. In addition, we find no strong evidence that suggests credit constraints lead to slower productivity growth. Finally, we confirm that both investment and employment growth are negatively affected by the measured credit constraints.

Birth Cohort Size Variation and the Estimation of Class Size Effects

We show that in school systems with grade retention or redshirting, birth cohort size is negatively related to the grade-level share of students who are too old for their grade. This compositional effect gives rise to an upward bias in estimates of class size effects based on commonly used research designs exploiting within-school variation in birth cohort size. Using data for all primary schools in one federal state of Germany, we find support for this compositional effect. Correcting for the resulting bias, we find that not only are smaller classes beneficial for test scores, but also for reducing grade repetitions.

Sequence effects in time trade-off valuation of hypothetical health states.

Choice-based stated preference methods, such as time trade-offs (TTOs), are used to establish health state utilities informing healthcare allocation. However, little is known about the presence of (position-dependent and precedent-dependent) sequence effects in the valuation of health states, despite techniques requiring respondents to evaluate several health states in a sequence. This paper is the first to explicitly test for the presence of sequence effects in the health domain using a new explanation based on contrast effects and preference imprecision; the implication being that randomisation cannot avoid sequence effects. Six TTO questions were designed using the EQ-5D-3L descriptive system. These were grouped into two blocks of three and within each block four sequences were used. In an online survey, 1,197 Spanish respondents answered one grouping of three TTO questions. Results indicate that sequence effects can affect preferences as utilities of health states are biased downwards if preceded by a better health state and biased upwards if preceded by a worse health state. This study informs our understanding of how context effects interact with preference elicitation methods, which is essential for interpreting survey results used to inform policy.

Bias in the estimation of cumulative viremia in cohort studies of HIV-infected individuals

PurposeThe use of cumulative measures of exposure to raised HIV viral load (viremia copy-years) is increasingly common in HIV prevention and treatment epidemiology due to the association of long-term elevated viral load with more rapid progression to disease. We sought to estimate the magnitude and direction of bias in a cumulative measure of viremia caused by different frequency of sampling and duration of follow-up. MethodsWe simulated longitudinal viral load measures and reanalyzed cohort study data sets with longitudinal viral load measurements under different sampling strategies to estimate cumulative viremia. ResultsIn both simulated and observed data, estimates of cumulative viremia by the trapezoidal rule show systematic upward bias when there are fewer sampling time points and/or increased duration between sampling time points, compared with estimation of full time series. Absolute values of cumulative viremia vary appreciably by the patterns of viral load over time, even after adjustment for total duration of follow-up. ConclusionsSampling bias due to differential frequency of sampling appears extensive and of meaningful magnitude in measures of cumulative viremia. Cumulative measures of viremia should be used only in studies with sufficient frequency of viral load measures and always as relative measures.

Revisiting dispersion in count data item response theory models: The Conway-Maxwell-Poisson counts model.

Count data naturally arise in several areas of cognitive ability testing, such as processing speed, memory, verbal fluency, and divergent thinking. Contemporary count data item response theory models, however, are not flexible enough, especially to account for over- and underdispersion at the same time. For example, the Rasch Poisson counts model (RPCM) assumes equidispersion (conditional mean and variance coincide) which is often violated in empirical data. This work introduces the Conway-Maxwell-Poisson counts model (CMPCM) that can handle underdispersion (variance lower than the mean), equidispersion, and overdispersion (variance larger than the mean) in general and specifically at the item level. A simulation study revealed satisfactory parameter recovery at moderate sample sizes and mostly unbiased standard errors for the proposed estimation approach. In addition, plausible empirical reliability estimates resulted, while those based on the RPCM were biased downwards (underdispersion) and biased upwards (overdispersion) when the simulation model deviated from equidispersion. Finally, verbal fluency data were analysed and the CMPCM with item-specific dispersion parameters fitted the data best. Dispersion parameter estimates indicated underdispersion for three out of four items. Overall, these findings indicate the feasibility and importance of the suggested flexible count data modelling approach.

Appraisal Accuracy and Automated Valuation Models in Rural Areas

Accurate and unbiased property value estimates are essential to credit risk management. Along with loan amount, they determine a mortgage’s loan-to-value ratio, which captures the degree of homeowner equity and is a key determinant of borrower credit risk. For home purchases, lenders generally require an independent appraisal, which, in addition to a home’s sales price, is used to calculate a value for the underlying collateral. A number of empirical studies have shown that property appraisals tend to be biased upwards, and over 90 percent of the time, either confirm or exceed the associated contract price. Our data suggest that appraisal bias is particularly pervasive in rural areas where over 25 percent of rural properties are appraised at more than five percent above contract price. Given this significant upward bias, we examine a host of alternate valuation techniques to more accurately estimate rural property values.

Controls, belief updating, and bias in medical RCTs

We develop a formal model of placebo effects. If subjects in seemingly-ideal single-stage RCTs update beliefs about breakthroughs based upon personal physiological responses, mental effects differ across medications received, treatment versus control. Consequently, the average cross-arm health difference becomes a biased estimator. Constructively, we show: bias can be altered through choice of control; higher-efficacy controls mitigate upward bias; and efficacy states can be revealed through controls of intermediate efficacy or controls that mimic a subset of efficacy states. Consistent with experimental evidence, our theory implies outcomes within-arm and cross-arm differences can be non-monotone in treatment probability. Finally, we develop novel differences-in-differences and triangle equality tests to detect RCT bias.

Measuring China's patent quality: Development and validation of ISR indices

Because China has become one of the largest applicants of PCT patents, it is of interest to compare the quality of Chinese and non-Chinese applications. We extend a quality index based on internationally comparable citation data from international search reports (ISR) to consider foreign, domestic, and self citations. Whereas foreign citations show that Chinese PCT patent applications reach only a third of the non-Chinese quality benchmark, the extension towards domestic and self citations suggests a higher quality level that converges to or even surpasses the benchmark. We investigate these differences based on firm-level regressions and find that in China, only foreign citations, but not domestic and self citations, have a significant and positive relation to R&D stocks. Using Germany as a representative country without policy support for patenting, we show that all three citation types may be used as economic indicators if policy distortion is not a concern. Our results show that domestic and self citations suffer from an upward bias in China and should be employed with caution if they are to be interpreted as a measure of patent quality.