The ING Group (Dutch: ING Groep) is a Dutch multinational banking and financial services corporation headquartered in Amsterdam. Its primary businesses are retail banking, direct banking, commercial banking, investment banking, wholesale banking, private banking, asset management, and insurance services. With total assets of US$1.1 trillion, it is one of the biggest banks in the world, and consistently ranks among the top 30 largest banks globally. It is among the top ten largest European companies by revenue.ING is the Dutch member of the Inter-Alpha Group of Banks, a cooperative consortium of 11 prominent European banks. Since the creation in 2012, ING Bank is a member in the list of global systemically important banks.In 2020, ING had 53.2 million clients in more than 40 countries. The company is a component of the Euro Stoxx 50 stock market index. The long-term debt for the company as of December 2019 is €150 billion.ING is an abbreviation for Internationale Nederlanden Groep (lit. 'International Netherlands Group'). The orange lion on ING's logo alludes to the group's Dutch origins.
Cathy Scott General Manager and Publisher In implementing a strategic asset allocation policy, the forecasting of long-term equity market returns is critically important. Although, historically, several econometric models have been employed for forecasting, more recently machine learning methods have been used for that purpose. In “The Best of Both Worlds: Forecasting US Equity Market Returns Using a Hybrid Machine Learning–Time Series Approach,” Haifeng Wang, Harshdeep Singh Ahluwalia, Roger A. Aliaga-Díaz, and Joseph H. Davis explore machine learning methods to forecast 10-year-ahead US stock returns. To compare the relative performance of machine learning methods, the authors compare the accuracy of these methods to the forecasts of one of the most commonly used regression-based forecasting models by asset managers, the traditional Shiller cyclically adjusted price-to-earnings (CAPE) ratio model. The authors find that machine learning techniques can only modestly improve the forecast accuracy of the regression-based CAPE ratio model. Moreover, they actually result in worse performance than a vector autoregressive model (VAR)–based two-step approach introduced in 2018 by three of the authors of this article. However, when the authors implement a hybrid ML-VAR approach (i.e., VAR-based two-step approach with machine learning techniques allowing for unspecified nonlinear relationships), they find up to 56% improvement in real-time forecast accuracy for 10-year annualized US stock returns. They find the ensemble method consistently offers the best out-of-sample forecast. Machine learning applications in finance have shown benefits over traditional linear models in forecasting stock returns. Edward Leung, Harald Lohre, David Mischlich, Yifei Shea, and Maximilian Stroh quantify these benefits by comparing the forecasting performance of commonly used machine learning algorithms with that of traditional linear methods in their article “The Promises and Pitfalls of Machine Learning for Predicting Stock Returns.” Using well-known equity characteristics, the authors forecast returns for largeand mid-cap stocks from various regional indexes using a gradient boosting machine (GBM) algorithm and standard ordinary least squares (OLS) approaches. In doing so, they shed light on the mechanics and results of the GBM model in order to alleviate its black-box character. While the forecasts from GBM models outperform OLS models based on statistical tests of forecasting performance, the economic gains from such nonlinear models depend on the ability to take the appropriate risks and efficiently implement trades. Jochen Papenbrock, Peter Schwendner, Markus Jaeger, and Stephan Krügel in their article “Matrix Evolutions: Synthetic Correlations and Explainable Machine Learning for Constructing Robust Investment Portfolios” use evolutionary algorithms to simulate correlation matrixes useful for financial market applications. Referring to their novel approach to generate realistic correlation matrixes as “matrix evolutions,” they explain how it can be used for many applications in asset management, such as generating risk scenarios for portfolios, pricing of multi-asset derivatives, backtesting investment strategies, and hedging correlation-dependent investment strategies and financial products. The potential application of matrix evolutions is demonstrated by the authors in a machine learning case study for robust portfolio construction in a multi-asset universe that shows how an explainable machine learning program b y gu es t o n Ju ne 1 5, 2 02 1. C op yr ig ht 2 02 1 Pa ge an t M ed ia L td .
_ Most of the largest oil and gas companies, along with many service providers, have set their sights on achieving net-zero Scope 1 emissions within the next 15 to 20 years. But when it comes to executing this ambition, the industry faces a simple but intimidating question: Where to begin? Greenhouse gases (GHGs) are woven into every stage of a well’s life which presents challenges that demand creative solutions that do not require too much capital. At the 2024 Abu Dhabi International Petroleum Conference (ADIPEC), a global lineup of operators and service providers shared their answers. More than a dozen technical papers on the topic of decarbonizing operations showed that tackling Scope 1 emissions—those that come directly from the extraction and production process—requires both technological ingenuity and a willingness to rethink longstanding practices. Houston area-based Canrig Technologies has developed a utility power module that allows any rig—not only the ones it owns—to tap into grid power. The innovation, outlined in SPE 221820, can cut emissions by an average of 20 metric tons of CO₂e per day when paired with LED overhead lighting. And in regions where grid power is cheapest, the subsidiary of Nabors Industries said the new module helps save up to 70% on fuel costs compared with running generators that guzzle 2,000 gal/D of diesel. Canrig isn’t alone in its effort to tamp down on drilling emissions. Rival drilling contractor Helmerich & Payne (H&P) presented its own solution at the conference in SPE 221821. Tulsa, Oklahoma-based H&P’s study showed fumigated dual-fuel engines that burn a mix of natural gas and diesel can slash rig emissions by 50% when compared with traditional generators. Yet, there is a tradeoff, which is why H&P’s engineers favor port-injected engines, despite offering only an 18% reduction compared to Tier 2 diesel generators. While this is less than half of the emissions drop achievable with the fumigated dual-fuel engines, H&P said the port-injected variant is the “most balanced approach” when it came to cost and reliability. The H&P team also found that port-injected engines, while achieving substantial CO₂e reductions, produced slightly more nitrogen oxides (NOx) than a conventional generator. NOx is not considered a GHG in the industry’s emissions tallies. Still, it’s a critical point to consider since NOx contributes to ozone formation and therefore indirectly fuels global warming. H&P’s authors acknowledged the discrepancy in their paper. “As a community, when continuously discussing GHG improvements and the cost associated with it, it is important to consider the incentives being signaled to vendors and technology providers,” they wrote about the tradeoffs between lower CO₂e and higher NOx. “Is this the intended signal? There is no clear answer, and this should be an active conversation.” Cement is often not the first thing that comes to mind in the push for decarbonization, yet the cement industry alone accounts for around 8% of global CO₂ emissions. And as a key ingredient of the well construction process, cement is now getting a hard look. The Abu Dhabi National Oil Company (ADNOC), in partnership with Halliburton, recently demonstrated how change is possible in this niche area. As highlighted in SPE 221838, the two companies replaced more than 70% of traditional Portland cement in a particularly challenging section of an offshore well with an alternative material. The result was a 70% reduction in cement-related emissions for that section of the well. Driven by its own emissions goals and tight regulatory requirements, Shell engineers came up with a new way to simulate the well cleanup and test phase while flaring as little natural gas as possible. When put into practice, the new methodology achieved an emissions output of just 17.5% of the permitted allowance and 30% of what similar startups typically emitted. Shell’s authors added that rig time was slashed to 45% of the expected duration for comparable operations—showing that the battle against emissions can yield win-win outcomes. More details on the new Shell methodology can be found in SPE 221841. Of course, the need to address flaring doesn’t end offshore, and Japanese producer INPEX has an idea: replace traditional flare tips with incinerators. The company’s study, presented in SPE 221898, zeroes in on small-scale operations, which account for about 40% of the industry’s flaring emissions, and found that incinerators which boast a 99.8% combustion efficiency rate could reduce emissions by 18%. The impact becomes even more striking when compared to low-efficiency flares—those with a methane combustion rate of around 90%. In such cases, emissions could be cut by more than 40% through the adoption of incinerators. As these papers demonstrate, eliminating Scope 1 emissions isn’t about a single silver bullet—it’s about doing the detective work to uncover which solutions deliver the biggest impact for the investment. Technologies like low-carbon cement and high-efficiency incinerators show promise, but as the industry navigates the early days of this journey, it’s still uncertain which approaches will prove both transformative in reducing operational emissions and scalable across the sector. For Further Reading SPE 221898 - Case Study: Practical Solutions for Managing Small-Volume Flares in Community Areas: The Installation of an Incinerator by R. Matsuo, Y. Takada, and T. Jeanes, INPEX Corp. SPE 221838 - Decarbonization and Successful Zonal Isolation of a Challenging Well Using a High-Performance Low-Carbon Emission System Containing Less Than 30% Portland Mass by M. Iskander, A.J.C. De Barros, A.B. De Melo, and M.R. Benberber, ADNOC Offshore; and S. Gupta, N. Aghayarova and P. Manga, Halliburton. SPE 221821 - From Baseline to Breakthrough: Comprehensively Decoding Rig Power Solutions for Sustainability and Scalability by S. Auld, A. Wood, J. Brady, and S. Kern, Helmerich & Payne. SPE 221820 - Operations Reduce Scope One Emissions Through Rig Electrification and Energy-Efficient Lighting by B. Landry, C. Stopkoski, and B. Norton, Canrig Technologies. SPE 221841 - Well Clean-Up and Testing: Using Dynamic Simulations to Minimise Flaring Emissions and Optimise Well Testing by J.H. Ellepola, Shell Markets, Middle East, Ltd.; S.E. Scott, Shell UK Ltd.; and M. Kerem, Shell Global Solutions International B.V.
Shale producers have long sought to understand how fractures propagate underground, a fundamental question that shapes well spacing, completion strategies, and overall economics. A new study from Devon Energy (SPE 223547) aims to provide answers. It introduces the concept of fracture growth rate, which can “vary drastically” between formations, and examines the key geological and geomechanical factors driving those differences. Devon defines fracture growth rate by the area increased per the volume injected (ft2/bbl) to create the fracture. “We noticed that the Eagle Ford has much slower growth rates than in the Anadarko, and it could be up to a four- or five-times difference,” said Brendan Elliott, subsurface engineering manager at Devon Energy, speaking at the SPE Hydraulic Fracturing Technology Conference and Exhibition (HFTC) in Houston. According to the Oklahoma City-based firm, the variations in growth rate observed across thousands of fracture stages not only impact per-well economics but also dictate how many wells an operator can drill per section. “If you think about a barrel injected as a proxy for volume or frac size, in some areas of the world, you’re getting much less area per barrel,” Elliott said. “That’s important to understand, especially as we contextualize the differences across our plays.” He then presented data comparing fracture growth rates across Devon’s operating areas, emphasizing a key trend: “If you look at the plays with some of the fastest growth rates, those tend to be the ones with the least dense well spacing.” Elliott pointed to the Turner Sands formation in Wyoming’s Powder River Basin, where most 640-acre drilling sections today support only two to three wells. By comparison, in the Eagle Ford Shale, well spacing of 330 ft often allows for as many as 14 wells per section. Devon’s findings are the result of more than 5 years of diagnostic work using its proprietary technology, sealed wellbore pressure monitoring (SWPM), along with other corroborating methods such as fiber optics. What’s behind the striking differences reported by Devon? The firm’s extensive data-gathering effort shows that the two primary factors driving difference in fracture growth rate are the rock’s stress state and geomechanical barriers which represent the boundary between two distinct benches or formations. Horizontal stress magnitudes affect the extent to which fractures open. Additionally, the stress barriers between zones will influence vertical height and whether fractures propagate through different formations too quickly. Secondary natural fractures further affect fluid movement by controlling leakoff, as do the properties of in-situ fluids. Other important rock properties to consider include Young’s modulus. At HFTC, Elliott explained that in rocks with a high Young’s modulus, like those found in the Anadarko Basin, fractures tend to have very narrow apertures causing them to extend far more laterally than vertically. In contrast, the Delaware Basin in the Permian Basin has a lower Young’s modulus, which is one reason Devon has determined that fractures there can extend up to 4,000 ft laterally—or fracture tip to fracture tip—and up to 2,000 ft vertically in a 2:1 ratio.
Reelwell’s DualLink wired drillpipe is heading offshore for the first time, set to make its debut later this year in a campaign with Vår Energi on the Norwegian Continental Shelf. The Stavanger-based technology developer sees the project as more than a milestone—it’s a chance to shift the narrative around wired pipe, a technology first brought to the market by NOV more than 2 decades ago. Challenged by high costs and integration hurdles, earlier generations of wired pipe never quite found widespread adoption. But Reelwell says it’s ready to change that. The DualLink system delivers real-time telemetry and, in a notable evolution, brings high-wattage downhole power into play for the first time. In this Q&A with JPT, Svein Strømberg, Reelwell’s newly appointed CEO and 31-year veteran of Baker Hughes, unpacks the company’s strategy as it prepares to put its most advanced technology to work offshore Norway. Editor’s note: This interview has been edited for length and clarity. SS: DualLink delivers high-speed data from the bottomhole assembly (BHA) to surface and high power from surface to the downhole BHA, enabling real-time transmission of both data and electrical power. It transmits at rates up to 56 kilobits per second (kbps), which is a significant leap compared to conventional mud pulse systems that typically operate between 1 and 30 bps. That level of bandwidth opens the door to much more detailed and frequent tool updates. But the real differentiator is electrical power. Our wired pipe can deliver up to 500 W downhole at present, enabling tools and functionality that simply are not possible with telemetry alone. Whether it’s active flow control, tool actuation, or high-power sensors, having a reliable power source changes what’s technically feasible today. Right now, surface certification standards place some limits on how much power can be supplied, but the system itself is built to scale. It is fully redundant and engineered to support greater capacity as requirements evolve. And for many current applications, 500 W is already more than enough to drive meaningful gains in capability. In our conversations with leading exploration and production companies, both in Norway and the US, the data speed was compelling, but what really stood out was the power. That’s what they saw as the key to enabling the next generation of downhole tools, and it has played a major role in recent contract awards. DualLink isn’t just about faster data, it’s about unlocking a smarter, more capable downhole environment. More broadly, we’re entering a market that has long been defined by a single legacy provider. That kind of landscape can limit innovation. With new players emerging, including Reelwell, we believe the increased competition will help accelerate the adoption of wired-pipe technologies and expand what’s possible in drilling operations. SS: There’s been hesitation in the past as wired pipe was often seen as expensive, complex, and difficult to plan for, especially when decisions had to be made a year in advance. And for some, the value wasn’t clearly understood. But I’m confident our upcoming work offshore in Norway will demonstrate that the landscape is changing. The benefits are becoming too significant to ignore.
INTRODUCTION:Obsessive-compulsive disorder (OCD) is a neuropsychiatric disorder characterized by recurrent intrusive thoughts and ritualized behaviors, often aimed at reducing distress. OCD is heterogeneous in its presentation and many patients with OCD experience a variety of different symptoms throughout their course of illness. Efforts to understand symptom domains in OCD have typically identified three to five symptom domains, such as the domains of doubt/checking, contamination, superstitions/rituals, symmetry/hoarding, and taboo thoughts. Recent studies in the genetics of OCD have suggested a common OCD dimension may provide additional information above and beyond the previously identified symptom domains. Thus, we sought to test a hierarchical model of lifetime OCD symptoms and evaluate the utility of the inclusion of a common OCD dimension. METHODS:Participants included 999 individuals participating in the OCD Collaborative Genetics Study (OCGS) and an additional 2363 individuals participating in the OCD Genetic Association Study (OCGAS). We evaluated unidimensional, 5-factor, and hierarchical models of lifetime OCD symptom presentation using confirmatory factor analysis. RESULTS:Results suggested that the hierarchical model best fit the data. Further evaluation of these models using a Bayesian testlet response model showed that lifetime presence of specific OCD symptoms was differentially associated with lifetime OCD severity. Moreover, symptoms associated with greater lifetime severity were generally reported less frequently than symptoms present at lower levels of lifetime severity. Implications of these findings and future directions are discussed.