Upward Bias Research Articles

Discounted-Cash-Flow Value Estimates are Optimistic when Assumptions are Not

Discounted-cash-flow valuations are upward biased even when estimates of market discount rates and cash flow growth rates are unbiased, because both compounding and discounting are convex functions. The upward bias in the estimated value of a long-lived cash flow stream can exceed 25% under plausible assumptions regarding magnitudes of estimation errors, which can equate to upward bias in estimated net present values that exceed 100%. This bias may help to explain the value premium puzzle, observed takeover premia, long run returns after IPOs, and high equity valuations during periods of low real interest rates. A method of correcting value estimates for the bias is proposed.

Degenerative dementia with proximal radial and lower vertical spatial neglect

ABSTRACT A 75 year-old man had a two-year history of progressive memory loss, trouble with finances and getting lost. On examination, he scored 16/30 in MoCA test, noticeably impaired on the attentional tasks. His screening bloodtests werenormal. Brain imaging revealed hippocampal atrophy and bilaterallarge areas of leukoaraiosis below posterior parietal lobes. On vertical line bisection he revealed a large upward bias and on radial bisection, a distal bias. Degeneration of his posterior parietal cortex may have caused both the leukoaraiosis and vertical-radial neglect. Unawareness of portions of space can be a source of disability and cause injury. Therefore, patients with degenerative dementia, especially those with similar patterns of leukoaraiosis or parietal degeneration should be tested for vertical and radial forms of spatial neglect.

Feeling and Looking Down: Impact of Depressive Symptoms on the Allocation of Vertical Attention.

Emotions and moods are often expressed using verbal, vertical spatial metaphors (eg, "I'm feeling down"). The reason for using these vertical metaphors is unknown; however, when individuals experience depressive symptoms, they often become self-reflective and ruminative. These mental activities are associated with activation of the default network, including the parietal lobes. The parietal lobes are a critical component of the dorsal visual attentional network, which allocates attention downward. Therefore, activation of the default network may create a downward bias of vertical attention. To investigate whether depressive mood disposition, as measured by the number of depressive symptoms experienced by an individual, influences their allocation of vertical attention. We used the Hospital Anxiety and Depression Scale to evaluate depressive symptoms in 48 right-handed individuals with a mean age of 57.31 (±17.14) years. We also asked the participants to mark the center (bisect) of 24 vertical and 24 horizontal lines. Overall, the individuals deviated their bisections of vertical lines upward (M=2.66 mm; SD=3.85). A stepwise linear regression analysis revealed that the number of depressive symptoms an individual experienced was significantly associated with a reduction in the magnitude of this upward vertical bias (P=0.009; R=0.138), such that the number of depressive symptoms was associated with the reduction (lowering) of their upward bias. Depressive symptoms may be associated with a relative lowering of the vertical attentional bias, and this lowering may be related to increased activation of portions of the default network.

Inventory statistics meet big data: complications for estimating numbers of species

We point out complications inherent in biodiversity inventory metrics when applied to large-scale datasets. The number of units of inventory effort (e.g., days of inventory effort) in which a species is detected saturates, such that crucial numbers of detections of rare species approach zero. Any rare errors can then come to dominate species richness estimates, creating upward biases in estimates of species numbers. We document the problem via simulations of sampling from virtual biotas, illustrate its potential using a large empirical dataset (bird records from Cape May, NJ, USA), and outline the circumstances under which these problems may be expected to emerge.

The impact of labor market work and educational tracking on student educational outcomes: Evidence from Taiwan

This study empirically investigates how working while enrolled in high school affects educational outcomes, while accounting for self-selected educational tracking. Using a longitudinal survey of Taiwanese youth and exploiting the inter-zip-code geographic variations in youth-preferred industries, we find a negative effect of school-year work on educational achievement, and the negative marginal impact is much stronger for academic-track than for vocational-track students. An exogenous increase in school-year hours worked of 10 hours per week lowers college entrance scores by a 0.117 (0.083) standard deviation for academic-track (vocational-track) students. The negative impact of school-year work tends to be overstated if the endogeneity arising from educational tracking is not acknowledged—with upward biases as large as 16% and 30% for academic- and vocational-track students, respectively. Among subjects, math scores suffer most from working during the school year.

The left posterior cerebellum is involved in orienting attention along the mental number line: An online-TMS study

Although converging evidence suggests that the posterior cerebellum is involved in visuospatial functions and in the orienting of attention, a clear topography of cerebellar regions causally involved in the control of spatial attention is still missing. In this study, we aimed to shed light on this issue by using online neuronavigated transcranial magnetic stimulation (TMS) to temporarily interfere with posterior medial (Vermis lobule VII) and left lateral (Crus I/II) cerebellar activity during a task measuring visuospatial (landmark task, Experiment 1 and 2) and representational (number bisection task, Experiment 2) asymmetries in the orienting of attention. At baseline, participants showed attentional biases consistent with the literature, that is a leftward and upward bias with horizontal and vertical lines, respectively, and a leftward bias in number bisection. Critically, TMS over the left cerebellar hemisphere significantly counteracted pseudoneglect in the number bisection task, whilst not affecting attentional biases in the landmark task. In turn, TMS over the posterior vermis did not affect performance in either task. Taken together, our findings suggest that the left posterior cerebellar hemisphere (but not the posterior vermis) is a critical node of an extended brain network subtending the control of spatial attention, at least when attention needs to be allocated to an internal representational space and a certain degree of mental manipulation is required (as in the number bisection task).

Calculating the Lung Allocation Score Error for ECMO Patients

The use of extracorporeal membranous oxygenation (ECMO) as a bridge to lung transplantation has increased dramatically. However, the Lung Allocation Score (LAS) does not use a dedicated ECMO variable; instead, ECMO is proxied by mechanical ventilation with 100% oxygen. We aimed to determine if this approximation accurately reflects the transplantation urgency for ECMO patients. Lung transplant candidates and recipients greater than 12 years of age between 2008-2014 were selected from the SRTR database. For both the wait-list (WL) and post-transplant (PT) time periods, predicted survival functions were calculated using the LAS model and compared to observed survival functions estimated by Kaplan-Meier. Differences between predicted and observed WL survival and PT survival were calculated for candidates and recipients, stratified by ECMO support at listing or transplant. The errors in WL and PT survival were then combined to estimate the LAS error for each group. During the study period, 16,259 candidates were listed; 236 were on ECMO at listing. A total of 12,169 patients were transplanted; 363 were on ECMO at transplant. Non-ECMO candidates survived a mean of 336 days from listing compared to a LAS prediction of 328 days, leading to a WL error of 9 days. ECMO candidates survived a mean of 199 days from listing compared to a LAS prediction of 123 days, leading to a WL error of 77 days. Non-ECMO recipients survived a mean of 333 days from transplant compared to a LAS prediction of 315 days, leading to a PT error of 17 days. ECMO recipients survived a mean of 296 days from transplant compared to a LAS prediction of 248 days, leading to a PT error of 48 days (Fig 1). ECMO was associated with an overestimation of the LAS by 10 points, compared to a non-ECMO underestimation of 0.04 points. The LAS ECMO correction underestimates both WL and PT one-year survival, leading to an overall upward bias in the LAS compared to non-ECMO candidates. This suggests that ECMO candidates have a lower transplantation urgency than implied by their LAS.

Patient satisfaction and survey response in 717 hospital surveys in Switzerland: a cross-sectional study

BackgroundThe association between patient satisfaction and survey response is only partly understood. In this study, we describe the association between average satisfaction and survey response rate across hospital surveys, and model the association between satisfaction and propensity to respond for individual patients.MethodsSecondary analysis of patient responses (166′014 respondents) and of average satisfaction scores and response rates obtained in 717 annual patient satisfaction surveys conducted between 2011 and 2015 at 164 Swiss hospitals. The satisfaction score was the average of 5 items scored between 0 and 10. The association between satisfaction and response propensity in individuals was modeled as the function that predicted best the observed response rates across surveys.ResultsAmong the 717 surveys, response rates ranged from 16.1 to 80.0% (pooled average 49.8%), and average satisfaction scores ranged from 8.36 to 9.79 (pooled mean 9.15). At the survey level, the mean satisfaction score and response rate were correlated (r = 0.61). This correlation held for all subgroups of surveys, except for the 5 large university hospitals. The estimated individual response propensity function was “J-shaped”: the probability of responding was lowest (around 20%) for satisfaction scores between 3 and 7, increased sharply to about 70% for those maximally satisfied, and increased slightly for the least satisfied. Average satisfaction scores projected for 100% participation were lower than observed average scores.ConclusionsThe most satisfied patients were the most likely to participate in a post-hospitalization satisfaction survey. This tendency produces an upward bias in observed satisfaction scores, and a positive correlation between average satisfaction and response rate across surveys.

Effects of Patient Size and Radiation Dose on Iodine Quantification in Dual-Source Dual-Energy CT

The purpose of this study was to investigate the potential effects of patient size and radiation dose on the accuracy of iodine quantification using dual-source dual-energy computed tomography (CT). Three phantoms representing different patient sizes were constructed, containing iodine inserts with concentrations from 0 to 20 mg/ml. Dual-energy CT scans were performed at six dose levels from 2 to 30 mGy. Iodine concentrations were measured using a three-material-decomposition algorithm and their accuracy was assessed. In a small phantom, iodine quantification was accurate and consistent at all dose levels. In a medium phantom, minor underestimations were observed, and the results were consistent except at low dose. In the large phantom, more significant underestimation of iodine concentration was observed at higher doses (≥15 mGy), which was attributed to the beam-hardening effect. At lower doses, increasing upward bias was observed in the CT number, leading to significant overestimations of both iodine concentration and fat fraction, which was attributed to the photon-starvation effect. The severity of the latter effect was determined by mA instead of mAs, suggesting that the electronic noise, rather than the quantum noise, was responsible for the bias. Using higher kVp for the low-energy tube was found to alleviate these effects. Reliable iodine quantification can be achieved using dual-source CT, but the result can be affected by patient size and dose rate. In large patients, biases may occur due to the beam-hardening and the photon-starvation effects, in which case higher dose rate and higher kVp are recommended to minimize these effects.

Optimizing design to estimate genetic correlations between environments with common environmental effects.

Breeding programs for different species aim to improve performance by testing members of full-sib (FS) and half-sib (HS) families in different environments. When genotypes respond differently to changes in the environment, this is defined as genotype by environment (G × E) interaction. The presence of common environmental effects within families generates covariance between siblings, and these effects should be taken into account when estimating a genetic correlation. Therefore, an optimal design should be established to accurately estimate the genetic correlation between environments in the presence of common environmental effects. We used stochastic simulation to find the optimal population structure using a combination of FS and HS groups with different levels of common environmental effects. Results show that in a population with a constant population size of 2,000 individuals per environment, ignoring common environmental effects when they are present in the population will lead to an upward bias in the estimated genetic correlation of on average 0.3 when the true genetic correlation is 0.5. When no common environmental effects are present in the population, the lowest standard error (SE) of the estimated genetic correlation was observed with a mating ratio of one dam per sire, and 10 offspring per sire per environment. When common environmental effects are present in the population and are included in the model, the lowest SE is obtained with mating ratios of at least 5 dams per sire and with a minimum number of 10 offspring per sire per environment. We recommend that studies that aim to estimate the magnitude of G × E in pigs, chicken, and fish should acknowledge the potential presence of common environmental effects and adjust the mating ratio accordingly.

THE ECONOMICS OF LUCKINESS: HOW CRAZY ARE MALAYSIAN CHINESE TOWARDS CAR PLATE NUMBER “8”?

The Chinese society is well-known for its obsession towards number “8”. Eight is the luckiest number to the Chinese as its pronunciation resembles the meaning of prosperity, wealth or fortune. Prior research estimating the marginal effect of “8” on winning bid prices of car plate numbers attributes the positive impact of “8” on bid prices to superstitious value. However, the estimated positive effect could also be attributed to Veblen goods and future resale consideration. Thus, an upward bias might exist in prior research’s estimation. This study aimed to circumvent this issue by comparing the marginal effect of all “8” car plate numbers of Malaysian Chinese bidders and their non-Chinese counterparts who are neutral towards number “8”. Descriptive statistics results reveal that the majority of all “8” plates are won by Chinese bidders. However, regression results reveal that Chinese bidders have a lower estimated marginal effect of number eight than their non-Chinese counterparts. This implies that while number “8” is popular among the Chinese, this popularity does not translate into higher bidding prices. These findings suggest the roles of Veblen goods and future resale consideration should not be ignored in estimating the superstitious value in car plate numbers.

How do Banking Analysts behave around Unanticipated News? Evidence from Operational Risk Event Announcements

We study earnings per share (EPS) forecast revision and accuracy of banking analysts around operational risk event announcements in U.S. banks. We find that first announcements of operational risk events are more informative than their settlement announcements. Optimistic banking analysts revise their forecasts downward more aggressively around operational risk disclosures, thereby improving forecast accuracy. Career concerns of banking analysts cause an upward bias in forecast revision and deterioration in forecast accuracy only if the potential employer is a systemically important bank (SIB). We find consistent evidence linking competition among banking analysts with optimistic and inaccurate forecasts, which is consistent with analysts seeking to use inflated forecasts to curry favour and attract businesses to their brokerage house around the time of operational risk disclosures. Global settlement has no favourable impact on analyst forecast accuracy around operational risk event announcements. We find evidence supporting a materiality threshold of $10 million for the informativeness of operational risk event announcements in SIBs. Overall, our results shed light on optimism bias in banking analyst behaviour upon the arrival of unanticipated news.

Asymmetric Cost Behaviour and Earnings Quality in the European Context

This study investigates the relationship of asymmetric cost behaviour with earnings quality for European listed firms. We employ a sample that consists of 11,416 firm-year observations of European listed firms over the period 2005-2019 to explore the relationship of asymmetric cost behaviour with (i) the managerial incentives to meet earnings targets, (ii) conditional conservatism, (iii) the level of operating accruals, and (iv) earnings smoothing. Our empirical evidence indicates that the presence of managerial incentives to meet earnings targets decreases the intensity of SG&A cost stickiness. We also document that asymmetric timeliness estimates in the conditional conservatism estimation models have an upward bias unless these models do not control for asymmetric cost behaviour. Finally, it seems that the levels of operating accruals and earnings smoothing are more sensitive to sales decreases than sales increases.

Man vs. Machine Learning: The Term Structure of Earnings Expectations and Conditional Biases

We introduce a real-time measure of conditional biases in firms' earnings forecasts. The measure is defined as the difference between analysts' expectations and a statistically optimal unbiased machine-learning benchmark. Analysts' conditional expectations are, on average, biased upwards, and the bias increases in the forecast horizon. These biases are associated with negative cross-sectional return predictability, and the short legs of many anomalies contain firms with excessively optimistic earnings. Further, managers of companies with the greatest upward-biased earnings forecasts are more likely to issue stocks. Commonly-used linear earnings models do not work out-of-sample and are inferior to those provided by analysts.

High Water, No Marks? Biased Lending after Extreme Weather

Policymakers have put forward proposals to ensure that banks do not underestimate long-term risks from climate change. To examine how lenders account for extreme weather, we compare matched repeat mortgage and property transactions around a severe flood event in England in 2013–14. First, lender valuations do not ‘mark-to-market’ against local price declines. As a result valuations are biased upwards. Second, lenders do not offset this valuation bias by adjusting interest rates or loan amounts. Third, borrowers with low credit risk self-select into high flood risk areas. Overall, these results suggest that lenders do not track closely the impact of extreme weather ex‑post, and that public flood insurance programs may subsidise high income households.

Estimating the Value of Chinese Superstitions in Car Plate Number of '8': A Nonlinear Decomposition Analysis

Estimating the value of numerical superstitions using the hedonic regression method faces an empirical challenge to separate the superstition and non-superstition components. In Malaysia, the number of eight is a lucky number by the Chinese as its pronunciation carries the meaning of “prosperity”; while for the non-Chinese (Malay), it is a neutral number (non-superstitious). This provides a natural setting to use the nonlinear decomposition analysis to remove the non-superstition component from the superstition where the hedonic regression is not able to handle. Using a sample of car plate auction data in Malaysia, this paper estimates the value of superstition in number of eight. The value of superstition is estimated to be RM965 by the hedonic regression. The nonlinear decomposition shows that it is only around 37% to 45% of the estimated superstition value (RM965) can be attributed to superstition. This indicates that the estimation of previous studies on superstitious value of number of eight using the hedonic regression model might be subject to a substantial upward bias.

Recent Evidence on International Stock Market’s Overreaction

This paper examines the overreaction hypothesis on market indices for three- and five-year investment periods using end-of-month data from 49 Morgan Stanley Capital International indices from December 1970 to December 2018. The returns were computed as holding-period returns, instead of cumulative average returns, to avoid an upward bias. We found economically and statistically significant return reversals for both the three-year and five-year investment periods. When implemented in developed markets only, there is evidence that supports the overreaction hypothesis, although the excess returns are smaller than those observed in the whole sample. Not only did the losers outperform the winners, but the former were also less risky. Notwithstanding these results, the overreaction strategy is sensitive to the periods considered, thus highlighting the possibility that its success is not time stationary.

Voluntary and Mandatory Social Distancing: Evidence on COVID-19 Exposure Rates from Chinese Provinces and Selected Countries

This paper estimates time-varying COVID-19 reproduction numbers worldwide solely based on the number of reported infected cases, allowing for under-reporting. Estimation is based on a moment condition that can be derived from an agent-based stochastic network model of COVID-19 transmission. The outcomes in terms of the reproduction number and the trajectory of per-capita cases through the end of 2020 are very diverse. The reproduction number depends on the transmission rate and the proportion of susceptible population, or the herd immunity effect. Changes in the transmission rate depend on changes in the behavior of the virus, reflecting mutations and vaccinations, and changes in people's behavior, reflecting voluntary or government mandated isolation. Over our sample period, neither mutation nor vaccination are major factors, so one can attribute variation in the transmission rate to variations in behavior. Evidence based on panel data models explaining transmission rates for nine European countries indicates that the diversity of outcomes results from the non-linear interaction of mandatory containment measures, voluntary precautionary isolation, and the economic incentives that governments provided to support isolation. These effects are precisely estimated and robust to various assumptions. As a result, countries with seemingly different social distancing policies achieved quite similar outcomes in terms of the reproduction number. These results imply that ignoring the voluntary component of social distancing could introduce an upward bias in the estimates of the effects of lock-downs and support policies on the transmission rates. The full set of estimation results and the replication package are available on the authors' websites.

The Macroeconomic Consequences of Infrastructure Investment

Can greater investment in infrastructure raise long-run output? Are infrastructure projects a good short-run stimulus to the economy? This paper uses insights from the macroeconomics literature to address these questions. I begin by analyzing the effects of government investment in both a stylized neoclassical model and a medium-scale New Keynesian model, highlighting the economic mechanisms that govern the strength of the short-run and long-run impacts. The analysis confirms earlier findings that the implementation delays inherent in infrastructure projects reduce short-run multipliers in most cases. In contrast, long-run multipliers can be sizable when government capital is productive. Moreover, these multipliers are greater if the economy starts from a point below the socially optimal amount of public capital. Turning to empirical estimation, I use the theoretical model to explain the econometric challenges to estimating the elasticity of output to public infrastructure. Using both artificial data generated by simulations of the model and extensions of existing empirical work, I demonstrate how both general equilibrium effects and optimal choice of public capital are likely to impart upward biases to output elasticity estimates. Finally, I review and extend some empirical estimates of the short-run effects, focusing on infrastructure spending in the ARRA.

Bayesian VAR Forecasts, Survey Information and Structural Change in the Euro Area

We incorporate external information extracted from the European Central Bank’s Survey of Professional Forecasters into the predictions of a Bayesian VAR, using entropic tilting and soft conditioning. The resulting conditional forecasts significantly improve the plain BVAR point and density forecasts. Importantly, we do not restrict the forecasts at a specific quarterly horizon, but their possible paths over several horizons jointly, as the survey information comes in the form of one- and two-year-ahead expectations. Besides improving the accuracy of the variable that we target, the spillover effects on “other-than-targeted” variables are relevant in size and statistically significant. We document that the baseline BVAR exhibits an upward bias for GDP growth after the financial crisis, and our results provide evidence that survey forecasts can help mitigate the effects of structural breaks on the forecasting performance of a popular macroeconometric model. Furthermore, we provide evidence of unstable VAR dynamics, especially during and after the recent Great Recession.