This research evaluates the efficacy of survival models in forecasting startup failures and investigates their economic implications. Several machine learning survival models, including Kernel SVM, DeepSurv, Survival Random Forest, and MTLR, are assessed using the concordance index (C-index) as a measure of prediction accuracy. The findings reveal that more sophisticated models, such as Multi-Task Logical Regression (MTLR) and Random Forest, outperform the standard Cox and Kaplan Meier (K-M) models in terms of predicted accuracy.

This study investigates the likelihood of time to bank failures in the US between 2001 and April 2023, based on data collected from the Federal Deposit Insurance Corporation's report on "Bank Failures in Brief - Summary 2001 through 2023". The dataset includes 564 instances of bank failures and several variables that may be related to the likelihood of such events, such as asset amount, deposit amount, ADR, deposit level, asset level, inflation rate, short-term interest rates, bank reserves, and GDP growth rate. We explore the efficacy of machine learning survival models in predicting bank failures and compare the performance of different models. Our findings shed light on the factors that may influence the probability of bank failures with a time perspective and provide insights for improving risk management practices in the banking industry.

Machine learning is playing an increasingly important role in the healthcare industry by transforming the way cancer is diagnosed and treated. By analyzing patient data, genomic data, and imaging data, machine learning algorithms can identify molecular signatures that distinguish cancer patients from healthy patients. Biomarkers that can accurately detect and diagnose cancer can be identified through analysis of these data sources. Additionally, personalized cancer therapies can be developed by identifying the most effective treatments based on individual patient characteristics and cancer type. Some of the machine learning techniques used for cancer biomarker discovery include deep learning and support vector machines, which can respectively identify complex patterns in data and classify data to identify relevant biomarkers. The benefits of using machine learning for cancer biomarker discovery are significant, including more precise and personalized treatments, improved patient outcomes, and the potential to transform cancer diagnosis and treatment. However, there are also challenges associated with using machine learning for cancer biomarker discovery, such as data collection and privacy issues, as well as the need for more powerful computational resources. This article explores the potential of machine learning in cancer biomarker discovery and argues that ongoing research in this field has the potential to revolutionize cancer diagnosis and treatment. Future research directions should focus on further developing machine learning algorithms and effective data collection and privacy protocols.

This paper explores how the medical expenditure risk affects the households’ portfolio choice across health status theoretically in a life cycle model and empirically using machine learning methods. Medical expenditure risk, as a background risk, has the potential to influence households’ financial decisions. A higher medical expenditure risk leads to a larger fluctuation and more uncertainty in households’ consumption and therefore utility. As a result, risk-free assets become more attractive. Our machine learning analysis provides evidence that aligns with the predictions of the theoretical life cycle model. Specifically, households with better health hold a larger proportion of stocks in their portfolios. Furthermore, when facing increased medical expenditure risk, households in good health demonstrate a greater willingness to invest in safe assets.

This research delves into the temporal dynamics of a nation's pursuit of a targeted GDP per capita level, employing five different survival machine learning models, remarkably Deep Learning algorithm (DeepSurv) and Survival Random Forest. This nuanced perspective moves beyond static evaluations, providing a comprehensive understanding of the developmental processes shaping economic trajectories over time. The economic implications underscore the intricate balance required between calculated risk-taking and strategic vulnerability mitigation. These findings guide policymakers in formulating resilient economic strategies for sustained development and growth amid the complexities inherent in contemporary economic landscapes.

We present two approaches to forecasting parameters in the SABR model. The first approach is the vector autoregressive moving-average model (VARMA) for the time series of the in-sample calibrated parameters, and the second is based on machine learning techniques called epsilon-support vector regression (ε-SVR). Using daily data of S&P 500 ETF option prices from January 1, 2014, to December 31, 2018, we first calibrate the daily values of the model parameters from the training sample, then conduct out-of-sample forecasting of parameters and pricing of options. Both approaches produce good fits between the forecasted and calibrated parameters for out-of-sample dates. A comparison study shows that using forecasted parameters as inputs, the SABR model generates better pricing results than assuming constant parameters or using lag parameters. We also discuss the market conditions under which one approach outperforms the other.

This paper provides a thorough review of the shifting landscape of economic analysis, spotlighting recent trends and predicting future paths. While traditional economic models remain key for interpreting economic activity, they are being supplemented by fresh methods and cross-disciplinary viewpoints. The increased attention to inequality studies, using advanced statistical techniques and unique data sources, underscores the growing emphasis on fairness and distribution within economic analysis. The incorporation of behavioral elements into economic models also expands our comprehension of economic decision-making and market results. Notably, the emergence of computational economics-integrating artificial intelligence (AI), big data, and machine learning into economic scrutiny-represents a major development. Often referred to as ’smart economics,’ this field employs technology to formulate, address complex economic dilemmas, and perceive economic activity in unconventional ways. Yet, the application of AI and machine learning in economics introduces new hurdles around data privacy, algorithmic bias, and the transparency of model outcomes. The impact of the digital revolution on economic analysis is significant, as the advent of computational economics and the surge of big data are transforming research techniques and policy implications. Concurrently, the advent of the circular economy indicates a radical shift in our perspective on economic sustainability, carrying considerable implications for environmental policy and business tactics.  In the future, it’s anticipated that these trends will further modify the realm of economic analysis, with AI and machine learning integration, emphasis on sustainability and fairness, and the influence of big data becoming more pronounced. As these changes take place, it’s imperative for researchers, policymakers, and practitioners to remain adaptable and flexible, prepared to capitalize on the opportunities and tackle the challenges these trends present.

Parkinson’s Disease (PD) is a prevalent progressive neurodegenerative condition affecting millions globally. Research has found that individuals with PD have a reduced risk of certain cancers, such as colon, lung, and rectal cancers, but an increased risk of brain cancer. Therefore, there is an urgent need for the development of advanced PD diagnostic methods and for investigating the relationships between risk factors, such as lifestyle due to handedness associated with various types of cancers. Recent ad- vancements in magnetic resonance imaging have enhanced PD diagnosis, reducing misdiagnosis and facilitating more accurate disease progression monitoring. Nevertheless, challenges exist, particularly in the distinction of PD between left-handed and right-handed patients over time. This survey provides an overview of contemporary deep learning-based imag- ing analysis methodologies, encompassing both non-longitudinal and lon- gitudinal contexts. We also explore existing limitations and prospects for refinement to gain deeper insights. These insights are poised to inform the development of personalized treatment strategies for PD patients while elucidating the current disparities between deep learning models and their efficacious implementation in clinical practice.

With the advancement of science and technology and the advent of the digital economy, the digital economy has become a new driving force for economic development, and the digital economy industry at home and abroad is facing new opportunities for survival and development. However, due to the special development model and profit mode of the digital economy industry, the traditional performance evaluation method is no longer applicable, so it is particularly important to improve the previous performance evaluation method. This paper first adjusts the customer value assessment (CLV) model according to the characteristics of the digital economy industry, and then analyzes the traditional EVA based on the adjusted user value assessment (ULV) model of the digital economy industry. The performance evaluation system is improved, and the representative enterprises of Alibaba, JD and Pinduoduo in the digital economy industry are selected for performance evaluation and analysis, and the user value performance evaluation model of the digital economy industry is integrated, innovated and standardized, in order to provide reference and reference for the follow-up research and practice of performance evaluation in the digital economy industry.

The use of cloud computing by firms can be expected to go hand in hand with higher productivity, more innovations, and lower costs, and, therefore, should be positively related to export activities. Empirical evidence on the link between cloud computing and exports, however, is missing. This paper uses firm level data for manufacturing enterprises from the 27 member countries of the European Union taken from the Flash Eurobarometer 486 survey conducted in February – May 2020 to investigate this link. Applying standard parametric econometric models and a new machine-learning estimator, Kernel-Regularized Least Squares (KRLS), we find that firms which use cloud computing do more often export, do more often export to various destinations all over the world, and do export to more different destinations. The estimated cloud computing premium for extensive margins of exports is statistically highly significant after controlling for firm size, firm age, patents, and country. Furthermore, the size of this premium can be considered to be large. Extensive margins of exports and the use of cloud computing are positively related.

In this work we aim to identify potential determinants and seek to predict terrorism attack. Thus, to eliminate uncertainty linked to explanatory variables we used the BMA method. We show that, contrary to expectations terrorism in MENA region is no longer purely of economic origin but mainly due to political problems, education, financial development and countries’ demographic characteristics. Likewise, we find that national, international and global terrorism are not of same origins even they present many common roots. In the end, we show that it is possible to predict majority of attacks based on a small number of indicators measuring political risk, financial development and income inequalities.

The dawn of the Artificial Intelligence (AI) era presents a plethora of new possibilities for analyzing regional economic development. The present article provides an in-depth exploration of the methods employed in this field, highlighting the immense opportunities that AI offers while also addressing potential challenges. The role of AI is crucial in complex data handling, enabling efficient analyses of intricate regional economic patterns. This capacity is paramount in shaping economic policies and strategies that are reflective of each region's unique needs and potential. The article firstly explores various AI methods used in economic analysis, including but not limited to machine learning, deep learning, and natural language processing. It delves into the application of these methods in discerning development trends, predicting economic shifts, and identifying strategic economic drivers unique to various regions. Subsequently, the potential of AI to transform regional economic analysis is discussed, encompassing its capability to process large and complex datasets, its power to predict future trends based on past and present data, and its ability to aid in strategic decision-making. However, this new era of AI-driven economic analysis is not without challenges. The latter part of this article thus confronts the issues related to data privacy, ethical use of AI, and the necessity of interdisciplinary skills in AI and economics. This exploration contributes to a broader understanding of how AI is transforming the landscape of regional economic development analysis, illuminating both its present use and future implications. By understanding these dynamics, we can better harness the potential of AI to advance economic prosperity in various regions around the globe.

Background: Bladder cancer (BLCA), particularly invasive BLCA, has become a medical burden worldwide as it is associated with recurrence and easy metastasis. There are specific differences in the expression of various miRNAs in tumor and normal tissues. Hence, miRNAs can be used as biomarkers for tumor diagnosis and prognostic evaluation. The current study aimed to predict the downstream target genes of BLCA-related miRNAs and explore their association with immune infiltration. Method: Data on BLCA-related mRNA and miRNA expression levels were downloaded from The Cancer Genome Atlas. Correlation analysis and Cox regression analysis were performed to validate the miRNA risk model. The infiltration of various immune cells should be compared to determine the distinct differences between the immunological microenvironment of the two risk groups. Results: A predictive framework of BLCA was established using the expression levels of two miRNAs. Cox regression analysis showed that the low-risk group had a better prognosis. Then, the target genes of miRNA were predicted, and the target genes were analyzed using Gene Ontology and Kyoto Encyclopedia of Genes and Genomes. Moreover, variations in immune cells and functions between the high- and low-risk groups were assessed. Conclusion: The prognostic features composed of two associated miRNAs (MIR-25, MIR-548AN) may help predict the overall survival of BLCA.

The latent transforming growth factor-beta (TGF-β) binding protein 1 (LTBP1) has been implicated in various cellular processes, but its role in triple-negative breast cancer (TNBC) remains unclear. In this study, we investigated the impact of LTBP1 on TNBC progression and its underlying mechanisms. Analysis of online datasets revealed elevated LTBP1 mRNA expression in breast cancer tissues compared to normal adjacent tissues. Kaplan-Meier Plotter analysis indicated that high LTBP1 expression was negatively correlated with relapse-free survival (RFS), distant-metastasis free survival (DMFS), and overall survival (OS) of breast cancer patients. Additionally, LTBP1 mRNA levels were associated with chemotherapy resistance. Functional assays in TNBC cells demonstrated that LTBP1 knockdown suppressed cell proliferation, induced apoptosis, and attenuated migration and invasion. In vivo studies confirmed that LTBP1 knockdown inhibited tumor growth in a xenograft mouse model. Mechanistically, LTBP1 positively correlated with genes involved in signaling regulation and organelle organization, with significant associations to GTPase binding and the RhoA/ROCK pathway. LTBP1 knockdown reduced RhoA activity and phosphorylation of Myosin Light Chain 2 (MLC2), suggesting inhibition of the RhoA/ROCK signaling pathway. Moreover, activation of the RhoA/ROCK pathway partially rescued the effects of LTBP1 knockdown on TNBC cell proliferation, apoptosis, migration, and invasion. In conclusion, our findings suggest that LTBP1 promotes TNBC progression through activation of the RhoA/ROCK signaling pathway, highlighting its potential as a therapeutic target for TNBC.

It's critical for environmental governance to understand the effectiveness of policy interventions as well as their working mechanisms. This paper focuses on the vehicle license plate-based management, a sort of policy intervention understudied in environmental governance literature. Specifically, we study the automobile purchase restriction (APR), a major method of license management to address urban air pollution and traffic congestion recently launched by the authorities of China. However, limited studies have quantified the environmental impact of such purchase restriction. Based on the panel data of PM2.5 concentration in 330 cities in China during 1998-2016, this study aims to explore the causal impact of Beijing automobile purchase restriction on air quality, using the regression control method and high-dimensional data regression of machine learning. The average PM2.5 concentration did not decrease but increased because of the purchase restriction. Over time, the deterioration effect on urban air quality will not alleviate. Besides, the restriction cannot enhance urban air quality by decreasing the increase of PM2.5. Despite excluding the impact of driving restriction, the purchase restriction markedly aggravate urban air quality. This paper has some policy implications for policymakers on the impact of urban vehicle license management with caution. Hence, it is essential to promote urban economy, production, and lifestyle to attain the win-win goal of air pollution control and urban development.

Background: Hepatocellular carcinoma (HCC) stands out as one of the most lethal cancers globally, given its complexity, recurrence following surgical resection, metastatic potential, and inherent heterogeneity. In recent years, researchers have systematically elucidated the significance of long non-coding RNA (lncRNA) in the initiation and progression of HCC. The introduction of The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases has significantly enhanced the prognostic assessment of HCC. However, the association between HCC and cell senescence has been infrequently explored in the literature. Method: We downloaded liver hepatocellular carcinoma (LIHC)-related messenger RNA and lncRNA expression levels from TCGA. Correlation analysis, Cox regression, and least absolute shrinkage and selection operator (LASSO) regression analysis were employed to validate the lncRNA risk model associated with cellular aging. Comparing the infiltration of diverse immune cells enabled the identification of distinct differences in the immunological microenvironments of the two risk groups. Subsequently, we conducted a real-time polymerase chain reaction (qPCR) experiment to confirm the accuracy of the selected lncRNAs. Results: A predictive framework for HCC was constructed based on the expression levels of five lncRNAs. Multivariate and univariate Cox regression analyses revealed that lncRNA signatures associated with senescence were independently correlated with an increased risk of HCC. Additionally, the nomogram also provides a more refined and sensitive model. Further investigation into the variations in immune cells and functions between the high-risk and low-risk groups was conducted. Subsequently, a qPCR experiment results revealed underexpression of AC068756.1, AC090578.1, AC145343.1, and LINC0022 in Huh7 and LM3 cells. In contrast, AP003392.4 did not exhibit a significant difference between Huh7 and control cells. Conclusion: The prognostic features and nomogram, consisting of five aging-related lncRNAs (AC068756.1, AC090578.1, AC145343.1, AP003392.4, and LINC00221), may be useful in predicting the overall survival of HCC.

This article discusses the pressing need to integrate artificial intelligence (AI) into education to facilitate customizable, individualized, and on-demand learning pathways. At the same time, while AI has the potential to expand the learner base and improve learning outcomes, the development of NACE Competencies and durable skills – communication, critical thinking, creativity, leadership, adaptability, and emotional intelligence - must be purposefully integrated in curriculum design now more than ever. Recent studies have shown that AI-driven learning pathways can achieve outcomes more quickly, but this comes at the cost of the development of durable skills. Therefore, traditional student-to-student and student-to-teacher interactions must be prioritized. As such, this study proposes a balanced approach to curriculum design to ensure the best outcomes for learners, where durable skill development is prioritized alongside subject-specific skills and rote memorization. Additionally, the article highlights the need for a combination of Just in Time Training (JITT) approaches, facilitated by AI technology, to reach the implementation of durable skills. The article concludes by questioning how to develop human skills in an increasingly AI-driven education system and emphasizes the importance of curriculum design and traditional learning approaches in creating a cohesive learning experience that develops durable skills in students. It is necessary to recognize that AI-driven education cannot replace the development of human skills, and that traditional interactions play a crucial role in developing these skills.

This paper attempts to introduce an overlapping generations structure into Paul Krugman's "The world's smallest macroeconomic model" (Krugman (1999)) to examine the implications of fiscal policy, particularly fiscal deficits, in a framework suitable for policy analysis. In that paper, Krugman argued that under the price rigidity assumption, a shortage in the money supply leads to underemployment and recession, so increasing the money supply would eliminate underemployment and restore full employment. But how can the money supply be increased? I show that in order to restore full employment out of a recession, a fiscal deficit is needed to increase the money supply. I also show that in a growing economy, fiscal deficits are necessary to maintain full employment at constant prices or inflation. Fiscal deficits are not only effective in pulling the economy out of recession, they are even necessary for growth to continue without recession or inflation. The fiscal deficit in this paper represents the difference between government spending and government revenues. If this difference is positive, we say that the government is in deficit. Krugman's original model is a one-period static model. I intend to extend this model to a dynamic overlapping generations model.

Vector Error Correction Models (VECM) have become a standard tool in empirical economics for analyzing nonstationary time series data because they integrate two key concepts in economics: equilibrium and dynamic adjustment in a single model. The current standard VECM procedure is limited to time series data with the same degree of integration, i.e., all I(1) variables. However, empirical studies often involve time series data with different de‐grees of integration, necessitating the simultaneous handling of I(1) and I(0) time series. This paper extends the standard VECM to accommodate mixed I(1) and I(0) variables. The conditions for the mixed VECM are derived, and consequently, we present a test and estimation for the mixed VECM.

In his "The World’s smallest macroeconomic model” (Krugman (1999)), Paul Krugman argued that under the assumption of price rigidity, a shortage of money supply leads to underemployment or recession, so increasing money supply can eliminate underemployment and restore full employment. But, how do we increase the money supply? I will show that we need a government deficit to increase the money supply in order to restore full employment from recession. Also, I will show that in a growing economy, if people hold money, a government deficit is necessary to maintain full employment under constant price or inflation. A government deficit is not only effective in pulling the economy out of recession, it is even necessary for continued growth without inviting either recession or inflation. The government deficit in this paper represents the difference between government expenditures and government revenues. When the difference is positive, we say that the government has a deficit. This paper seeks to explore theoretically and normatively the role of government deficits in achieving and maintaining full employment in a growing economy without causing inflation, using a very simple model by Krugman.