Historical Average Research Articles

Tool for tracking all-cause mortality and estimating excess mortality to support the COVID-19 pandemic response.

ProblemQuantifying mortality from coronavirus disease (COVID-19) is difficult, especially in countries with limited resources. Comparing mortality data between countries is also challenging, owing to differences in methods for reporting mortality.ContextTracking all-cause mortality (ACM) and comparing it with expected ACM from pre-pandemic data can provide an estimate of the overall burden of mortality related to the COVID-19 pandemic and support public health decision-making. This study validated an ACM calculator to estimate excess mortality during the COVID-19 pandemic.ActionThe ACM calculator was developed as a tool for computing expected ACM and excess mortality at national and subnational levels. It was developed using R statistical software, was based on a previously described model that used non-parametric negative binomial regression and was piloted in several countries. Goodness-of-fit was validated by forecasting 2019 mortality from 2015–2018 data.OutcomeThree key lessons were identified from piloting the tool: using the calculator to compare reported provisional ACM with expected ACM can avoid potential false conclusions from comparing with historical averages alone; using disaggregated data at the subnational level can detect excess mortality by avoiding dilution of total numbers at the national level; and interpretation of results should consider system-related performance indicators.DiscussionTimely tracking of ACM to estimate excess mortality is important for the response to COVID-19. The calculator can provide countries with a way to analyse and visualize ACM and excess mortality at national and subnational levels.

Compound impact of land use and extreme climate on the 2020 fire record of the Brazilian Pantanal

AbstractAimOur aim was to quantify the influence of climate and land use on major fires that occurred during the 2020 drought over the Brazilian Pantanal region.LocationAlto Paraguay Basin, central‐western flank of Brazil.Time period2003–2020.MethodsWe calculated climatic and burned area anomalies and Spearman's correlation between precipitation and sea surface temperature (SST). We assessed water coverage to identify the impact of the 2020 drought. We produced fire recurrence maps, identified the areas that burned for the first time in 2020, and estimated the carbon loss. Lastly, we assessed the burned areas within land‐use classes and land tenure.ResultsThe anomalies showed that climatic variables in 2020 had an atypical behaviour, being hotter and drier than the historical average. The water surface area in the Pantanal has been decreasing annually from 2003, and in 2020 the reduction was 34% greater than the annual average. Burned areas in 2020 were 200% greater than the long‐term average, and 35% of the burned areas occurred for the first time in this year. We showed that 84% of these new fires occurred within natural vegetation, with 39% of the burned areas occurring in forests, an increase of 514%. Forest fires were responsible for 47% of the carbon loss in 2020. Finally, 70% of the fires in 2020 occurred within rural properties, 5% in indigenous lands and 10% in protected areas.Main conclusionsThe unprecedented fires in 2020 were not only a result of atypical climate, but also a consequence of the intensification of fire‐related human activities. Rural properties play a major role in Pantanal fires, requiring law enforcement and a consistent plan for increasing the protection of the conservation areas. We argue that ineffective fire prevention and the lack of suppression strategies were also key in enhancing the magnitude of these fires.

Multi-Purpose, Multi-Step Deep Learning Framework for Network-Level Traffic Flow Prediction

This study proposes a data fusion and deep learning (DL) framework that learns high-level traffic features from network-level images to predict large-scale, multi-route, speed and volume of connected vehicles (CVs). We present a scalable and parallel method of processing statewide CVs’ trajectory data that leads to real-time insights on the micro-scale in time and space (two-dimensional (2D) arrays) on graphics processing unit (GPUs) using the Nvidia rapids framework and dask parallel cluster, which provided a 50× speed-up in the data extraction, transform and load (ETL). A UNet model is then applied to perform feature extraction and multi-route speed and volume channels over a multi-step prediction horizon. The accuracy and robustness of the proposed model are evaluated by taking different road types, times of day and image snippets and comparing the model to benchmarks: Convolutional Long–Short-Term Memory (ConvLSTM) and a historical average (HA). The results show that the proposed model outperforms benchmarks with an average improvement of 15% over ConvLSTM and 65% over the HA. Comparing the image snippets from each prediction model to the actual image shows that image textures were highly similar in UNet to the benchmark models used. UNet’s dominance in performing image predictions was also evident in multi-step forecasting, where the increase in errors was relatively minimal over longer prediction horizons.

Extreme temperatures during pregnancy and adverse birth outcomes: Evidence from 2009 to 2018 U.S. national birth data

We provide the first estimates of the impacts of prenatal exposure to extreme temperatures on infant health at birth using the latest national birth data from 2009 to 2018 from all U.S. states. We consistently find that an additional day with mean temperature greater than 80°F or less than 10°F increases preterm births and low birthweight. Strikingly, the adverse effects are borne disproportionately by Black and Hispanic mothers, suggesting that the projected increase in extreme temperatures may further exacerbate the existing birth health disparities across different race/ethnicity groups. We also contribute by investigating the impact of deviations from the normal weather pattern, to identify the extreme weather events after accounting for the adaptation response. We find that prenatal exposure to extreme heat two standard deviations above county's historic average induces preterm births and NICU admissions, particularly for mothers whose pregnancies overlap with summer months. These results are timely and policy relevant, considering the recent weather trends with rising temperatures and frequent extreme weather events.

Climate services for sustainable resource management: The water—energy—land nexus in the Tărlung river basin (Romania)

Decision-makers need tailored information regarding future climate, land use, forest management and societal scenarios for sustainable watershed management. Such information can be attained by integrating the water-energy-land nexus approach with climate services. To support decision-makers from Brașov metropolitan area (Romania), we co-developed land use, forest management and societal scenarios and analysed their impact on the water-energy-land nexus under climate change. We applied the Soil and Water Assessment Tool (SWAT) to the Săcele reservoir from the Tărlung river basin in order to spot the interrelations between nexus components under different climate and societal scenarios. In particular, we applied four bias-corrected GCM & RCM model combinations for two climate scenarios RCP4.5 and RCP8.5, nine land use and forest management scenarios, and twelve societal scenarios. Modelling results were analysed between 2020 and 2099. The results for the study area indicate that the projected average annual precipitation is close to historical averages in most climate scenarios explored, while the projected average annual temperature is increasing. Furthermore, the analysed data suggest an increase in severe dry intervals from approximately 6% (for the period 1961–2013) to 12% by the end of the 21st century, indicating that precipitation variability may change considerably. The SWAT model results show no significant changes in the water flow under different forest management scenarios. In contrast, model results for scenarios that imply forests' clearcutting and their replacement with pastures or meadows indicate that water quality is affected by increased sediment flow. Regarding the number of months with water scarcity in the Săcele reservoir (i.e. water demand exceeds water supply), we found an increase from 0.4 months in 2020 to 1.5 months in 2099. The model results also revealed that climate change significantly influences the water supply in the reservoir. To conclude, the regional stakeholders highly value the model results for the co-developed scenarios, which have been already used to promote sustainable resource management.

Forecasting equity risk premium: A new method based on wavelet de‐noising

AbstractForecasting equity risk premium is notoriously difficult due to a mass of data noise in the raw series and the absence of clear tendency. Using the monthly S&P 500 excess returns from 1927:01 to 2018:12, we first de‐noise the in‐sample original returns series via wavelet method to capture the basic trend of equity risk premium, and then propose forecasting models to obtain one‐step forward out‐of‐sample predicted values based on the de‐noised returns. Our new models can provide substantially superior out‐of‐sample performance compared with other competing models and the historical average. Sizeable economic gains can be realized by a mean–variance investor if they allocate their asset through the new approach. Our findings are robust under different settings from both statistical and economic perspectives.

Does previous exposure to extreme precipitation regimes result in acclimated grassland communities?

Climate change will likely increase weather persistence in the mid-latitudes, resulting in precipitation regimes (PR) with longer dry and wet periods compared to historic averages. This could affect terrestrial ecosystems substantially through the increased occurrence of repeated, prolonged drought and water logging conditions. Climate history is an important determinant of ecosystem responses to consecutive environmental extremes, through direct damage, community restructuring as well as morphological and physiological acclimation in species or individuals. However, it is unclear how community restructuring and individual metabolic acclimation effects interact to determine ecosystem responses to subsequent climate extremes. Here, we investigated, if and how, differences in exposure to extreme or historically normal PR induced long-lasting (i.e. legacy) effects at the level of community (e.g., species composition), plant (e.g., biomass), and molecular composition (e.g., sugars, lipids, stress markers). Experimental grassland communities were exposed to long (extreme) or short (historically normal) dry/wet cycles in year 1 (Y1), followed by exposure to an identical PR or the opposite PR in year 2 (Y2). Results indicate that exposure to extreme PR in Y1, reduced diversity but induced apparent acclimation effects in all climate scenarios, stimulating biomass (higher productivity and structural sugar content) in Y2. In contrast, plants pre-exposed to normal PR, showed more activated stress responses (higher proline and antioxidants) under extreme PR in Y2. Overall, Y1 acclimation effects were strongest in the dominant grasses, indicating comparatively high phenotypical plasticity. However, Y2 drought intensity also correlated with grass productivity and structural sugar findings, suggesting that responses to short-term soil water deficits contributed to the observed patterns. Interactions between different legacy effects are discussed. We conclude that more extreme PR will likely alter diversity in the short-to midterm and select for acclimated grassland communities with increased productivity and attenuated molecular stress responses under future climate regimes.

Examining Historical Rates of Leafcutting Bee Brood Cell Pathogens, Parasitoids, and Predators to Establish Baseline Infectivity Rates for Alfalfa Seed Growers.

The alfalfa leafcutting bee Megachile rotundata (Fabricius) (Hymenoptera: Megachilidae) is the primary pollinator for the alfalfa seed industry. It is a solitary cavity nesting bee that utilizes leaf lined brood cells provisioned with pollen for larval development and pupation into the adult stage. During development, multiple pathogens, parasitoids, and predators can prey upon or use the larvae as a host, resulting in the loss of the future adult bee. As such, the presence of invertebrate pests and fungal pathogens presents a major concern for commercial alfalfa seed growers. In the present study, we used historic data from the Parma Cocoon Diagnostic Laboratory to determine baseline rates of pathogens, parasitoids, and predators of Megachile rotundata brood cells and used this analysis to determine cutoffs for management practices to inform growers when the purchase of new bee stocks should be considered. Additionally, we compared the presence of chalkbrood, predators, and parasitoids in samples collected from both grower-produced stocks and newly purchased Canadian bees. The results of the investigation provide historic averages of the presence of chalkbrood, predators, and parasitoids, show a significant increase in chalkbrood and predators in 2007–2011, and find a significant difference in rates of chalkbrood and predators between samples from Canadian suppliers and grower stocks. We speculate that these differences may have resulted from economic conditions that increased the cost of Canadian Megachile rotundata cells and likely resulted in increased reliance on 2nd-year U.S. grower stocks and subsequently increased infection rates during this time period.

Joint prediction of zone-based and OD-based passenger demands with a novel generative adversarial network

Online ride-hailing plays an important role in modern urban transportation systems, and accurate short-term passenger demand prediction contributes to improving ride-hailing services. Many existing studies have made great achievements in zone-based demand prediction. However, origin–destination (OD)-based demand prediction has attracted little attention, even though it provides abundant information and is of high practical importance, e.g., it facilitates the routing and matching of ride-hailing services. Considering the importance of both types of demands, a conditional generative adversarial network (CGAN) with Wasserstein divergence (CWGAN-div) is proposed to jointly predict zone-based and OD-based demands. Residual blocks are utilized to capture internal spatiotemporal features, which benefit the training and prediction processes of the CWGAN-div model. Conditional information is also incorporated to characterize the external dependencies of the demands. Numerical experiments are performed by using GPS trajectory data from Didi Chuxing, Chengdu, China. The results show that the proposed CWGAN-div model yields good joint prediction performance and outperforms both classic models (historical average (HA) and convolutional neural network (CNN) models) and other prevailing GAN models (i.e., GAN, CGAN, Wasserstein GAN with a gradient penalty (WGAN-GP), and GAN with Wasserstein divergence (WGAN-div)). The proposed CWGAN-div model displays promise for the joint prediction of zone-based and OD-based passenger demands.

The predictability of cross-sectional returns in high frequency

Stock return forecast is of great importance to trading, hedging, and portfolio management. In this article, we apply LASSO and random forest to make rolling one-minute-ahead return forecasts of Dow Jones stocks, using the cross-section of lagged returns of S&P 500 components as candidate predictors. Although the number of candidate variables is large, the negative out-of-sample R2 suggests that the predictions from LASSO and random forest give larger mean-squared error than the historical average. So, there is no evidence of predictability in the cross-sectional returns of large stocks in high frequency. The predictability presented by Chinco et al. (2019) might be due to the interaction between large and small stocks.

Drinking Water Disinfection Byproducts, Ingested Nitrate, and Risk of Endometrial Cancer in Postmenopausal Women.

Background:Disinfection byproducts (DBPs) and N-nitroso compounds (NOC), formed endogenously after nitrate ingestion, are suspected endometrial carcinogens, but epidemiological studies are limited.Objectives:We investigated the relationship of these exposures with endometrial cancer risk in a large prospective cohort.Methods:Among postmenopausal women in the Iowa Women’s Health Study cohort, we evaluated two major classes of DBPs, total trihalomethanes (TTHM) and five haloacetic acids (HAA5), and nitrate-nitrogen () in public water supplies (PWS) in relation to incident primary endometrial cancer (1986–2014). For women using their PWS at enrollment (; ), we computed historical averages of annual concentrations; exposures were categorized into quantiles and when possible percentile. We also computed years of PWS use above one-half the U.S. maximum contaminant level (; TTHM; HAA5; ). Dietary nitrate/nitrite intakes were estimated from a food frequency questionnaire. We estimated hazard ratios (HR) and 95% confidence intervals (CI) via Cox models adjusted for age, endometrial cancer risk factors [e.g., body mass index, hormone replacement therapy (HRT)], and mutually adjusted for . We evaluated associations for low-grade () vs. high-grade () type I tumors. We assessed interactions between exposures and endometrial cancer risk factors and dietary factors influencing NOC formation.Results:Higher average concentrations of DBPs (95th percentile: TTHM , HAA5 ) were associated with endometrial cancer risk (TTHM: , 95% CI: 1.41, 3.40; HAA5: , 95% CI: 1.19, 2.83; ). Associations were similarly observed for women greater than median years of PWS use with levels , in comparison with zero years (TTHM: , 95% CI: 1.18, 2.21; HAA5: , 95% CI: 1.31, 2.62). Associations with DBPs appeared stronger for low-grade tumors (TTHM: , 95% CI: 1.17, 3.83; ) than for high-grade tumors (TTHM: , 95% CI: 0.80, 2.44; ), but differences were not statistically significant (). Associations with TTHM were stronger among ever HRT users than non-HRT users (). We observed no associations with in drinking water or diet.Discussion:We report novel associations between the highest DBP levels and endometrial cancer for our Iowa cohort that warrant future evaluation. https://doi.org/10.1289/EHP10207

The COVID-19 pandemic and associated rise in pediatric firearm injuries: A multi-institutional study

BackgroundFirearm sales in the United States (U.S.) markedly increased during the COVID-19 pandemic. Our objective was to determine if firearm injuries in children were associated with stay-at-home orders (SHO) during the COVID-19 pandemic. We hypothesized there would be an increase in pediatric firearm injuries during SHO. MethodsThis was a multi institutional, retrospective study of institutional trauma registries. Patients <18 years with traumatic injuries meeting National Trauma Data Bank (NTDB) criteria were included. A “COVID” cohort, defined as time from initiation of state SHO through September 30, 2020 was compared to “Historical” controls from an averaged period of corresponding dates in 2016–2019. An interrupted time series analysis (ITSA) was utilized to evaluate the association of the U.S. declaration of a national state of emergency with pediatric firearm injuries. ResultsNine Level I pediatric trauma centers were included, contributing 48,111 pediatric trauma patients, of which 1,090 patients (2.3%) suffered firearm injuries. There was a significant increase in the proportion of firearm injuries in the COVID cohort (COVID 3.04% vs. Historical 1.83%; p < 0.001). There was an increased cumulative burden of firearm injuries in 2020 compared to a historical average. ITSA showed an 87% increase in the observed rate of firearm injuries above expected after the declaration of a nationwide emergency (p < 0.001). ConclusionThe proportion of firearm injuries affecting children increased during the COVID-19 pandemic. The pandemic was associated with an increase in pediatric firearm injuries above expected rates based on historical patterns.

Cancer Prevention in a Postpandemic World: A One-Stop-Shop Approach

Cancer Prevention in a Postpandemic World: A One-Stop-Shop Approach

Estimating groundwater recharge in the assiniboine delta aquifer using HYDRUS-1D

Unconfined aquifers are often directly impacted by high irrigation demands. Therefore, accurate recharge estimates are crucial for sustainable groundwater management. It is important over time to re-examine the existing hydrologic understanding of aquifers, taking into account developments in knowledge and changing climate conditions. The last in depth study into the Assiniboine Delta Aquifer located in Manitoba, Canada, that gave a recharge estimate was conducted in the 1980s (Render, 1988). This work re-examined the Assiniboine Delta Aquifer, employing new methods and data to estimate recharge rates. Twelve one-dimensional models were created in the software HYDRUS-1D to model the soil water fluxes in the unsaturated zone to analyze the historic recharge from 1996 to 2019. Remote weather station data, measured hydraulic conductivity, soil texture distribution, soil moisture content, and soil temperature data were used to initialize and run the models. Inverse calibration used the measured moisture contents to calibrate the models, which resulted in Root Mean Square Error in the calibration and validation periods averaging 0.034 and 0.051 m3/m3, respectively, above the 0.025 m3/m3 sensor measurement error. Rough estimates of soil texture distribution across the aquifer was determined to assist in the final recharge estimate. The historical regional recharge average was estimated to be 68 mm/year, double the previous estimate for the aquifer (Render, 1988). Modelled periods were limited to less than a year due to limitations in the general HYDRUS-1D software model code under frozen soil conditions, which the Assiniboine Delta Aquifer experiences during the winter months. Assumptions on initial soil moisture contents, snowpack heights, and model start dates were best estimated for historical years. Results show these estimates can have significant impacts on the resulting recharge. Suggested future work includes implementing the HYDRUS-1D freeze-thaw code to allow for model spin-up and multi-year simulations to enhance the reliability of model results.

Precipitation Microphysics during the Extreme Meiyu Period in 2020

Previous studies have reported the large-scale meteorological conditions and dynamic causes of the extreme period of meiyu rainfall in 2020. However, the microphysical properties of meiyu precipitation during this period remain unclear. We used the Global Precipitation Measurement 2ADPR orbital precipitation dataset, the IMERG gridded precipitation dataset and the ERA5 reanalysis dataset to study the characteristics of meiyu precipitation over the Yangtze Plain during the extreme meiyu period in 2020 and historical meiyu periods from 2014 to 2019. The results showed that the average daily rainfall during the 2020 meiyu period was 1.5 times higher than the historical average as a result of the super-strong water vapor flux in the low- to mid-level layers of the atmosphere. The amplitude of nocturnal low-level water vapor transport during the 2020 meiyu period was twice the historical average and, therefore, the diurnal peak of meiyu rainfall at 0630 LST in 2020 was significantly earlier than the historical average. The moisture transport was the dominant moisture supply for precipitation during the 2020 meiyu period, whereas the moisture convection contributed less than in the meiyu periods of 2014–2019. This led to the precipitation in the 2020 meiyu period having a higher concentration of smaller droplets than the historical average. There were lower proportions of size-sorting evaporation and break-up processes in the liquid-phase precipitation processes in the 2020 meiyu than the historical average, but a higher proportion of coalescence processes. These results provide a factual basis for the simulation and forecast of precipitation during extreme meiyu periods.

Projected Changes in the Tibetan Plateau Snowpack Resulting From Rising Global Temperatures

AbstractThe snowpack is a critical component of the hydrologic cycle in cold regions, the change in which becomes important for proper planning and management of water. The Tibetan Plateau provides significant amount of water to most Asian rivers, and consequently the downstream population is dependent on its availability. Despite its importance, potential change in snowpack in this region due to climate change is poorly understood to date, largely because of remoteness and the orographic complexity of the area. This study inspects the impact of climate change on snowpack change over the Tibetan Plateau considering historical simulations (1981–2004), near future projections (2041–2064), and far future projections (2071–2094) from global climate models (GCMs) and regional climate models (RCMs) of derived temperature, precipitation, and snow water equivalent (SWE). A multivariate nesting bias correction approach (MRNBC) was employed to correct possible biases in GCM and RCM derived temperature, precipitation, and SWE jointly over multiple time scales to preserve interdependencies among the variables while enabling simulation of year‐to‐year persistence, which is of importance in water security assessments. MRNBC reduced the bias in model simulations significantly and improved projections of the snow climatology. The results indicate that the annual maximum spell of snow‐free days will increase over the region whereas the snowy day fraction will decrease in the future compared to the historical period. In addition, annual SWE are noted to be decreasing in both the near future and far future with respect to historical averages. Changes in SWE will result from warming temperatures and also from changes in precipitation, which will lead to more rainfall than snowfall thus affecting snowmelt processes.

Association of Economic Recession and Social Distancing With Pediatric Non-accidental Trauma During COVID-19.

IntroductionThere has been concern that the incidence of non-accidental trauma (NAT) cases in children would rise during the COVID-19 pandemic due to the combination of social isolation and economic depression. Our goal was to evaluate NAT incidence and severity during the pandemic across multiple US cities.MethodsMulti-institutional, retrospective cohort study comparing NAT rates in children <18 y old during the COVID-19 pandemic (March-August 2020) with a recent historical data (January 2015-February 2020) and during a previous economic recession (January 2007-December 2011) at level 1 Pediatric Trauma Centers. Comparisons were made in local and national macroeconomic indicators.ResultsOverall rates of NAT during March-August 2020 did not increase compared to historical data (P = 0.8). Severity of injuries did not increase during the pandemic as measured by Glasgow Coma Scale (GCS) (P = 0.97) or mortality (P = 0.7), but Injury Severity Score (ISS) slightly decreased (P = 0.018). Racial differences between time periods were seen, with increased proportions of NAT occurring in African-Americans during the pandemic (P < 0.001). NAT rates over time had low correlation (r = 0.32) with historical averages, suggesting a difference from previous years. Older children (≥3 y) had increased NAT rates during the pandemic. Overall NAT rates had low inverse correlation with unemployment (r = −0.37) and moderate inverse correlation with the stock market (r = −0.6). Significant variation between sites was observed.ConclusionsOverall NAT rates in children did not increase during the COVID-19 pandemic, but rates were highly variable by site and increases were seen in African-Americans and older children. Further studies are warranted to explore local influences on NAT rates.

Accessing Future Mixed Beans Yield in Zambia under a Changing Climate Scenario

Quantitative evaluation of the response of crop yield and crop water productivity to future climate change is important to prevent/mitigate the adverse effects of climate change. This study assessed the future impact of climate change on rain-fed mixed beans production in Zambia. Mixed beans yield data from 1976 to 2000, was used, aggregated by province and was detrended for non-climatic factors, then correlated with the climatic variables from WABAL function in MOSAICC programme. The significantly correlated variables were regressed to generate coefficients (xi) for each variable (ai) and the intercept (b) to build models for each province. Downscaled data from three GCMs of CanESM2, CNRM-CM5 and MPI-ESM-MR were used to forecast the 2011-2069 yields under RCP 4.5 and RCP 8.5. By comparing the historical and the future mean yields for each RCP, the change was computed and judged for significance. The general trend in the projected climate simulated future yields by the three GCMs under the two RCPs showed a decrease in yield in Agro-Ecological Region (AER) III, IIb and I, and an increase in AER IIa. The projected decrease was attributed to the temperature increase and rainy seasons tending to start later and end earlier than the historical average. This change calls for use of varieties that are more resilient to elevated mean temperatures, use of early maturing varieties, selection of suitable planting dates, the use of chemical fertilizers and the construction of irrigation and water conservancy facilities.

Attention-Driven Recurrent Imputation for Traffic Speed

In practice, traffic data collection is often warned by missing data due to communication errors, sensor failures, storage loss, among other factors, leading to impaired data collection and hampering the effectiveness of downstream applications. However, existing imputation approaches focus exclusively on estimating the lost value from incomplete observations and ignore historical data. In this paper, we propose a novel neural network model, namely, Attention-Driven Recurrent Imputation Network (ADRIN), to address the problem of missing traffic data. Specifically, in ADRIN, we devise an Imputation-targeted Long Short-Term Memory (LSTM-I) module for filling in missing data. Meanwhile, we consider the periodicity of historical data and design a historical average calculation module in ADRIN. On this basis, we employ the multi-head self-attention mechanism for further extracting latent temporal features from the output of the two modules. ADRIN is capable of modeling both incomplete observation inputs and historical averages independently to estimate the missing values. We conducted comprehensive experiments on three real-world traffic datasets, to demonstrate that ADRIN consistently outperforms other baselines in a variety of scenarios. Furthermore, ablation experiments are conducted on the various modules of the model, and it is concluded that historical data can significantly enhance the imputation effect.

Disruption and Credit Markets

In the past thirty years, defaults on corporate bonds have been substantially higher than the historical average. We show that this increase in credit risk can be largely attributed to an increase in the rate at which new and fast-growing firms displace incumbents (a phenomenon sometimes referred to as 'disruptive innovation'). Industries with a lager presence of firms newly listed on the stock market, as well as industries that receive funding from venture capital, have a higher loss of revenue market share for established firms and subsequently see a rise in corporate bond defaults. Patent filings by individuals as opposed to corporations also predict defaults. These results are not affected by inclusion of controls for industry exposure to offshore manufacturing.