Tuesday, September 01, 2020

The best of both worlds for economic predictions

by Ng Yi-Di, Singapore Managment University
Credit: CC0 Public Domain

Danish physicist Neils Bohr once quipped that prediction is hard, especially when it is about the future. But this is precisely what financial regulators need to do—forecasting the likely state of the economy in the future is crucial when deciding on policy levers like whether to slash or raise interest rates.

However, as the world continues to become more unpredictable, forecasting has become increasingly difficult. This challenge was poignantly illustrated after the start of the 2008 Financial Crisis, when Queen Elizabeth asked a seemingly simple but pointed question to a room of researchers and economists at the London School of Economics: Why did no one see it coming?

In the face of great complexity, perhaps econometrics could do with more help. Take machine learning for example. With its ability to parse big data, it could improve on existing econometric methods and lead to better forecasts. This is the research that Professor Yu Jun of the Singapore Management University (SMU) presented along with Associate Professor Xie Tian of the Shanghai University of Finance and Economics, at a webinar organized by SMU and the Monetary Authority of Singapore (MAS) on 26 June 2020.

In their talk titled "Econometric Methods and Data Science Techniques," Professors Yu and Xie reviewed existing econometric methods and machine learning techniques before discussing a hybrid of both methods. Using real data and examples, they showed that the hybrid method may herald better economic and financial variable forecasting.

"We all know that we are in the era of big data and machine learning data science techniques," said Professor Yu. "Some people may think that machine learning poses a threat to conventional econometric methods. Is that really the case?"


A mix of tradition and change

Speaking first, Professor Yu introduced a selection of traditional econometric methods. Blending economics with statistics, econometrics takes a structured quantitative statistical approach to economic analyses. With econometrics, Professor Yu explained, the method is to use past data to establish statistical relationships which in turn can be used to forecast possible futures.

"Most econometric methods hope to facilitate interpretation and statistical inference," he said, explaining that conventional econometric methods rely on assumptions and linear relationships, like the famous linearity assumption. "You want to map out from the past to the future."

This approach works well in certain cases, Professor Yu said, but also proves to be a limitation: most conventional econometric models cannot handle big data or complicated relationships. "If you have many predictors or a complicated relationship, econometric methods will fail. And that's a serious limitation in the big data era and in many important cases," he said.

On the other hand, Professor Xie, who is also an Adjunct Professor at SMU, explained that machine learning algorithms are data-driven. "Instead of relying on assumptions, many machine learning algorithms just let the data talk: they don't impose very strong assumptions or restrictions on the data-generating process," he said.

That's what makes machine learning techniques so flexible, Professor Xie pointed out. However, he also added that many machine learning methods are not truly tailored for economic and financial data in the first place.

So when it comes to trying to predict the future, is one approach better than the other? Like many things in this world, it's not so straightforward.

Weighing their strengths and weaknesses

Professors Yu and Xie used two real-world examples to illustrate how the two different approaches can outperform each other in accuracy depending on the data and case at hand.

In the forecasting of the Volatility Index, or VIX—an index of financial market volatility created by the Chicago Board Options Exchange—they showed that more traditional linear econometric modeling produced a more accurate forecast than more complex machine learning methods.

However, in the second case of forecasting the consumer price inflation of the eurozone, machine learning methods outperformed traditional econometric methods.

"Machine learning methods are very popular, but they do not always outperform conventional econometric methods. The question is, can we modify machine learning algorithms to adopt advanced econometric techniques and use economic data better?"

In this vein, Professors Yu and Xie discuss the idea that applying machine learning methods to existing econometrics approaches, instead of using either approach separately, could improve econometrics modeling. They suggest a hybrid algorithm, a model averaging regression tree (MART), which was first proposed by Professor Xie and Professor Steven F. Lehrer of Queen's University in a 2018 NBER Working Paper.

Putting MART to the test

To test the effectiveness of this hybrid method, they used it to forecast real economic and financial variables by applying it to the same examples of VIX and eurozone inflation rates discussed earlier.

Econometric models still had the best forecasting accuracy in forecasting VIX, performing better than their hybrid MART approach. Professors Yu and Xie suggest that this is because the VIX data exhibits very strong linearity and therefore is best suited for an econometric approach.

But when it came to forecasting the eurozone inflation rates, the results showed that the duo's hybrid approach performed best, generating superior forecast accuracy compared to either econometrics or machine learning methods alone.

So while a hybrid model isn't always the best in all cases, elements of machine learning may still improve on forecasting by picking up on trends that the traditional econometric models might miss. As for now, they say the key is to understand the fundamentals of each method and apply them in the most appropriate circumstances.

"The hybrid strategy combines econometric measures with machine learning strategies to lead to significant gains in forecasting accuracy," said Professor Xie. "Of course, this is just an idea. Future work is definitely needed to understand the properties of this proposed hybrid strategy in order to help guide practitioners."


Explore further  Computers excel in chemistry class
Provided by Singapore Management University
SWEDISH WORKERS BEST PAID IN EUROPE IN LATE FIN DE SICLE 19TH CENTURY

by Uppsala University
Credit: CC0 Public Domain

In 19th-century Sweden, workers' wages rose faster than in other European countries. By 1900, they were among the highest in Europe, and the steepest rise of all had been for those who earned least. This is shown by new research at Uppsala University: a study published in he Journal of Economic History.


"Historians often describe Sweden in the late 19th century as a poor country. Our results show the need for a more nuanced view. Although poverty did exist, of course, great changes were under way, and unskilled laborers seem to have been among those who benefited most from the upward trend."

The speaker, Johan Ericsson, is a researcher at the Department of History, Uppsala University. He and his colleagues have surveyed Swedish pay trends in the building and construction industry in the period 1831-1900. By using such sources as wage statistics from the Board of Public Buildings (the precursor to the National Property Board Sweden) and published research on the subject, they were able to compile figures on wages for four categories of construction workers: handymen, carpenters, bricklayers and the draft-horse drivers who transported the materials.

The results show that wage rates were increasing throughout Sweden, and pay differentials among occupational categories were shrinking. During the period, unskilled handymen's real wages rose most rapidly: by 176 per cent.

For corresponding workers in cities like Amsterdam, Antwerp, Paris and London, wages increased by between 40 and 90 per cent in the same period. This meant that the Swedish handymen's wages at the century's end were some 30 per cent higher than those of their counterparts in Paris, Amsterdam and Antwerp. Wages in London, which were the highest in Europe, were some 12 per cent higher than those of average laborers in Sweden.


This international comparison was feasible once the researchers had recalculated wages in terms of welfare ratios. In brief, this meant working out the quantity of certain products that an individual wage could purchase.

The researchers' conclusions are that a labor market with high mobility, combined with mass emigration to America that reduced the supply of unskilled labor, can explain why Swedish pay rates rose so rapidly.

"One intriguing observation is that workers' wages increased faster than average incomes in society. These days, there's a lot of talk about how globalization and technological development are making the workers' situation relatively worse. However, our findings show that this is no natural law. On the contrary, it seems that Swedish workers in this period were favored by trends like that," says Jakob Molinder of the Department of Economic History at Uppsala University.


Explore further Increase in immigration has little impact on the wages of US citizens

More information: Johan Ericsson et al, Economic Growth and the Development of Real Wages: Swedish Construction Workers' Wages in Comparative Perspective, 1831–1900, The Journal of Economic History (2020). DOI: 10.1017/S0022050720000285

Provided by Uppsala University

Monitoring and reporting framework to protect World Heritage Sites from invasive species

by CABI
Aldabra is among the world’s largest coral atolls but, like many World Heritage Sites, is at risk from invasive alien species. Credit: Adam Plezer, Seychelles Islands Foundation).

A team of international scientists have devised a new monitoring and reporting framework to help protect World Heritage Sites from almost 300 different invasive alien species globally including, rats (Rattus spp.), cats (Felis catus), lantana (Lantana camara) and Argentine ants (Linepithema humile).


Lead author Dr. Ross Shackleton joined invasive species experts from around the globe—including CABI's Dr. Arne Witt—who suggest the new 'tool' could ultimately help protect World Heritage Sites like the Galápagos, Serengeti and Aldabra Atoll from future threats.

UNESCO World Heritage Sites, are areas of outstanding universal value and conservation importance. However, they are threatened by a variety of global change drivers, including biological invasions from a range of terrestrial, freshwater and marine-based invasive alien species.

The team, who assessed biological invasions and their management in 241 natural and mixed World Heritage Sites from documents collated by UNESCO and the International Union for Conservation of Nature (IUCN), found that reports on the status of biological invasions were often inconsistent.

The research, published in Biodiversity and Conservation, outlined that while some reports were 'very informative', the scientists say they were hard to compare because no systematic method of reporting was followed.

Dr. Shackleton added, "Detailed information on invasive alien species management undertaken in World Heritage Sites was available for fewer than half of the sites that listed invasive alien species as a threat.

"There is clearly a need for an improved monitoring and reporting system for biological invasions in World Heritage Sites and likely the same for other protected areas globally."

The new framework devised by the scientists, which has been tested at seven World Heritage Sites, covers collecting information and reporting on; pathways, alien species present, impacts, management, predicting future threat and management needs, the status of knowledge and gaps, and, assigning an overall 'threat score' to the protected area.
The great Serengeti wildebeest migration is also at risk from a number of invasive alien plant species. Credit: Arne Witt

Dr. Witt said, "We need urgent action right now to reduce the severity of these threats that include a range of invasive alien plant species—such as Mimosa pigra and Prosopis juliflora—and we believe that the development of this monitoring and reporting framework is a step in the right direction to protecting areas moving forward."


Based upon a previous review of invasive plants in the Serengeti-Mara ecosystem, Dr. Witt added, "Failure to act could, for instance, see the devastation of the Serengeti-Mara ecosystem as we know it and that would have a major impact upon the annual wildebeest migration." Dr. Jäger concurs, "Invasive mammals and invertebrates in Galápagos threaten some of the animals made famous by Darwin, such as the giant tortoise and Darwin's finches."

In testing the devised framework, which categorises the level of threat posed by invasive alien species as 'very high', 'high', 'moderate' and 'low; the researchers have already yielded more information than from past monitoring initiatives.

For example, the invasive alien species threat level indicated in the 2017 IUCN World Heritage Outlook for the Serengeti, Keoladeo, Doñana and the Vredefort Dome sites was 'data deficient' or 'low threat' or 'not listed', whereas all of these World Heritage Sites are now categorised as facing 'moderate' to 'high' threats from biological invasions.

Co-author Prof. David Richardson said, "World Heritage Sites face growing threats from a range of biological invasions which impact upon native biodiversity and the delivery of ecosystem services. Not only that but invasive alien species are a financial burden as costs for management can be extremely high."

"One key element of the new framework is listing all invasive alien species present where we can track the changes in threat or implementation of effective management over time."

This is exemplified by work done on Aldabra Atoll where, according to co-author Dr. Nancy Bunbury, "There has been a decrease in the number of IAS listed due to effective eradications highlighting great management success at the site over the last few years."

The scientists, in their recommendations, also suggest that funding should be made available to conduct surveys at all under-resourced World Heritage Sites, to inform the reactive 'state of conservation' assessments undertaken by UNESCO and IUCN, and that monitoring could also be enhanced by members of the public as part of a series of 'citizen science' projects.


Explore further Protected areas worldwide at risk of invasive species

More information: Ross T. Shackleton et al, Biological invasions in World Heritage Sites: current status and a proposed monitoring and reporting framework, Biodiversity and Conservation (2020). DOI: 10.1007/s10531-020-02026-1
Climate change fuels sharp increase in glacier lakes

by Marlowe Hood   

SEPTEMBER 1, 2020

Unlike normal lakes, glacier lakes are unstable because they are often dammed by ice or sediment composed of loose rock and debris

The volume of lakes formed as glaciers worldwide melt due to climate change has jumped by 50 percent in 30 years, according to a new study based on satellite data.

"We have known that not all meltwater is making it into the oceans immediately," lead author Dan Shugar, a geomorphologist and associate professor at the University of Calgary, said in a statement.

"But until now there were no data to estimate how much was being stored in lakes or groundwater."

The findings, published Monday in Nature Climate Change, will help scientists and governments identify potential hazards to communities downstream of these often unstable lakes, he said.

They will also improve the accuracy of sea level rise estimates through better understanding of how—and how quickly—water shed by glaciers makes it to the sea.

Between 1994 and 2017, the world's glaciers, especially in high-mountain regions, shed about 6.5 trillion tonnes in mass, according to earlier research.

"In the past 100 years, 35 percent of global sea-level rises came from glacier melting," Anders Levermann, climate professor at the Potsdam Institute for Climate Change Impact, told AFP.

The other main sources of sea level rise are ice sheets and the expansion of ocean water as it warms.
A glacial lake at the end of the Rhone Glacier, near Gletsch on August 3, 2018

Glacial lake outbursts

Earth's average surface temperature has risen one degree Celsius since preindustrial times, but high-mountain regions around the world have warmed at twice that pace, accelerating glacier melt.

Unlike normal lakes, glacier lakes are unstable because they are often dammed by ice or sediment composed of loose rock and debris.

When accumulating water bursts through these accidental barriers, massive flooding can occur downstream.

Known as glacial lake outbursts, this kind of flooding has been responsible for thousands of deaths in the last century, as well as the destruction of villages, infrastructure and livestock, according to the study, published in Nature Climate Change.

The most recent recorded incident was a glacial lake outburst that washed through the Hunza Valley in Pakistan in May.
Between 1994 and 2017, the world's glaciers, especially in high-mountain regions, shed about 6.5 trillion tonnes in mass

In January, the UN Development Programme estimated that more than 3,000 glacial lakes have formed in the Hindu Kush-Himalayan region, with 33 posing an imminent threat that could impact as many as seven million people.

The new study, based on 250,000 scenes from NASA's Landsat satellite missions, estimates current glacial lake volume at more than 150 cubic kilometres (37 cubic miles), equivalent to one-third the volume of Lake Erie in the United States or twice the volume of Lake Geneva.

A decade ago, it would have not been possible to process and analyse that volume of data, said Shugar.


Explore furtherHimalayan lakes are exacerbating glacial melt

More information: Rapid worldwide growth of glacial lakes since 1990, Nature Climate Change (2020). DOI: 10.1038/s41558-020-0855-4

Journal information: Nature Climate Change



© 2020 AFP

Be generous, live longer

sharing
Credit: CC0 Public Domain
The act of giving and receiving increases well-being: the recipient benefits directly from the gift, and the giver benefits indirectly through emotional satisfaction. A new study published in the journal PNAS now suggests that those who share more also live longer. In their analysis, Fanny Kluge and Tobias Vogt found a strong linear relationship between a society's generosity and the average life expectancy of its members. The researchers at the Max Planck Institute for Demographic Research in Rostock, Germany, conclude that people are living longer in societies whose members support each other with resources.
"What is new about our study is that for the first time we have combined transfer payments from state and family and evaluated the effect", says Fanny Kluge. The researchers used data for 34 countries from the National Transfer Accounts project. For all countries, state and private transfer payments received and given by each individual over his or her lifetime are added up and presented in relation to lifetime income.
Societies in Western European countries share a lot and live long
Sub-Saharan African countries such as Senegal share the lowest percentage of their lifetime income and have the highest mortality rate of all the countries studied. Those who share little die earlier. Although South Africa is economically more developed than other African countries, few resources are redistributed; here too, the mortality rate is relatively high. In these countries, the mortality rate of children and  up to the age of 20 is also higher than in the other countries studied. "Our analyses suggest that redistribution influences the mortality rate of a country, regardless of the per capita gross domestic product," says Fanny Kluge.
Societies in Western European countries and Japan transfer a lot to the youngest and oldest and mortality rates are low. The countries studied in South America also have high transfer payments. There, people  more than 60 percent of their average life income with others. The mortality rates are lower than in sub-Saharan Africa, but higher than those of Western Europe, Australia, Japan and Taiwan.
In France and Japan, the two countries with the lowest  rates of all the countries studied, an average citizen shares between 68 and 69 percent of their lifetime income. Here, the risk of dying in the coming year is only half as high for people over 65 as in China or Turkey, where between 44 and 48 percent of lifetime income is redistributed.
"What I find particularly interesting is that the relationship between generosity and lifetime income that we described does not depend on whether the benefits come from the state or from the wider family," says Fanny Kluge. Both of these factors cause the population live longer compared to societies with fewer .

Explore further
Europe can fight virus without lockdowns: WHO

More information: Tobias Vogt et al, Intergenerational resource sharing and mortality in a global perspective, Proceedings of the National Academy of Sciences (2020). DOI: 10.1073/pnas.1920978117
Provided by Max Planck Society 
122 shares
Feedback to editors


Aluminum recycling technology boosted by crystallization research

Aluminium recycling technology boosted by crystallisation research
Under the influence of magnetic stirring, helical screw-like crystals form in molten alloy. Credit: Dr Biao Cai
An innovative method for aluminum recycling has been boosted by research showing the microscopic changes that take place when molten alloys cool.
Researcher Dr. Biao Cai from the University of Birmingham's School of Metallurgy and Materials used sophisticated high-speed X-ray imaging to record the formation of micro-crystals as alloys cool and solidify, under a .
 was developed by his collaborator Dr. Andrew Kao from the University of Greenwich to predict whether micro-crystals would form, and what shape they would have.
The model predicted that that helical 'screw-like' crystals would form under the influence of strong magnetic stirring, and the high-speed X-ray confirmed that this occurred.
Although these elegant crystals are just micro-meters wide (ten times smaller than a ), they have implications for industrial-scale processes.
Biao explains: "These microscopic crystals ultimately determine the physical properties of the alloy. To be able to adjust their shape, structure and direction of growth will enable us to perfect processes for both manufacturing and recycling of metals and alloys".
Biao has already invented technique to improve aluminum recycling by removing . Iron is a detrimental element that can make aluminum brittle, and limit its use in premium applications such as aircraft.
Existing methods for removing iron during recycling are either expensive or inefficient, but Biao's simple, inexpensive technique uses magnets and a  to remove iron contamination.Using AI to predict new materials with desired properties

More information: B. Cai et al. Revealing the mechanisms by which magneto-hydrodynamics disrupts solidification microstructures, Acta Materialia (2020). DOI: 10.1016/j.actamat.2020.06.041
Provided by University of Birmingham 

Workplace climate drives nurses' perception of burnout

Workplace climate drives nurses' perception of burnout
(HealthDay)—Workplace climate is the factor most associated with burnout in nurses, according to a study published Sept. 1 in the American Journal of Critical Care.
Lakshmana Swamy, M.D., from the Boston Medical Center, and colleagues evaluated the frequency of burnout and individual and organizational characteristics associated with burnout among critical care nurses across a national integrated health care system using data from an annual survey. Analysis included responses to the 2017 survey from 2,352 critical care nurses from 94 sites.
The researchers found that one-third of nurses reported burnout, which varied significantly across sites. Workplace climate was the strongest predictor of burnout (odds ratio [OR], 2.20), but other significant variables included overall hospital quality (OR, 1.44), urban location (OR, 1.93), and  tenure (OR, 2.11). The workplace climate subthemes of perceptions of workload and staffing, supervisors and senior leadership, culture of teamwork, and patient experience were each significantly associated with burnout.
"Drivers of  are varied, yet interventions frequently target only the individual," the authors write.

Explore further
Burnout high among non-physician frontline health care workers

More information: Abstract/Full Text
Copyright © 2020 HealthDay. All rights reserved.
 shares