Search results

This study aims to critically evaluate the competitiveness of Transformer-based models in financial forecasting, specifically in the context of stock realized volatility forecasting. It seeks to challenge and extend upon the assertions of Zeng et al. (2023) regarding the purported limitations of these models in handling temporal information in financial time series.

Employing a robust methodological framework, the study systematically compares a range of Transformer models, including first-generation and advanced iterations like Informer, Autoformer, and PatchTST, against benchmark models (HAR, NBEATSx, NHITS, and TimesNet). The evaluation encompasses 80 different stocks, four error metrics, four statistical tests, and three robustness tests designed to reflect diverse market conditions and data availability scenarios.

The research uncovers that while first-generation Transformer models, like TFT, underperform in financial forecasting, second-generation models like Informer, Autoformer, and PatchTST demonstrate remarkable efficacy, especially in scenarios characterized by limited historical data and market volatility. The study also highlights the nuanced performance of these models across different forecasting horizons and error metrics, showcasing their potential as robust tools in financial forecasting, which contradicts the findings of Zeng et al. (2023)

This paper contributes to the financial forecasting literature by providing a comprehensive analysis of the applicability of Transformer-based models in this domain. It offers new insights into the capabilities of these models, especially their adaptability to different market conditions and forecasting requirements, challenging the existing skepticism created by Zeng et al. (2023) about their utility in financial forecasting.

While different attempts have been made to use artificial intelligence (AI) to codify communicative behaviors and analyze startups’ video presentations in relation to crowdfunding projects, less is known about other forms of access to entrepreneurial finance, such as video pitches for candidacies into startup accelerators and incubators. This research seeks to demonstrate how AI can enable the startup selection process for both entrepreneurs and investors in terms of video pitch evaluation.

An AI startup (Speechannel) was used to predict the outcomes of startup video presentations by analyzing text, audio, and video data from 294 video pitches sent to a leading European startup accelerator (LUISS EnLabs). 7 investors were also interviewed in Silicon Valley to establish the differences between humans and machines.

This research proves that AI has profound implications with regards to the decision-making process related to fundraising and, in particular, the video pitches of startup accelerators and incubators. Successful entrepreneurs are confident (but not overconfident), engaging in terms of speaking quickly (but also clearly), and emotional (but not overemotional).

This study not only fills the existing research gap but also provides a practical guide on AI-driven video pitch evaluation for entrepreneurs and investors, reshaping the landscape of entrepreneurial finance thanks to AI. On the one hand, entrepreneurs could use this knowledge to modify their behaviors, enabling them to increase their likelihood of being financially backed. On the other hand, investors could use these insights to better rationalize their funding decisions, enabling them to select the most promising startups.

This paper makes a significant contribution by bridging the gap between theoretical research and the practical application of AI in entrepreneurial finance, marking a notable advancement in this field. At a theoretical level, it contributes to research on managerial decision-making processes – particularly those related to the analysis of video presentations in a fundraising context. At a practical level, it offers a model that we called the “AI-enabled video pitch evaluation”, which is used to extract features from the video pitches of startup accelerators and incubators and predict an entrepreneurial project’s success.

Social networks (SNs) play a significant role as environments supporting teacher professional development. The purpose of this to analyze the motivation and participation roles that Spanish teachers have when participating in SNs for their professional development in three professional stages: preservice teachers, beginning teachers and experienced teachers.

The study uses a mixed-method approach, combining two validated surveys, one applied to 217 preservice teachers and other to 68 beginning teachers and 384 experienced teachers, with 15 interviews. A qualitative exploratory sequential strategy has been followed along with an ex post facto quantitative survey-type study of a descriptive and inferential nature.

Preservice and beginning teachers use SNs to access materials and resources with which to learn, presenting an observer and passive role in their interaction on SNs. Experienced teachers log in to learn about experiences but begin to participate more actively in SNs for searching for specific resources, establishing contacts with other teachers, contributing with their own educational materials and helping other teachers with their doubts or even forming their own communities.

These findings help understand how the evolution in teacher expertise accompanies the level of involvement in their social network interactions. The results allow us to better understand how different levels of teaching experience influence the way Spanish teachers access and participate in SNs, in some cases consuming and in others producing digital content.

The purpose of the study was to gather data to determine whether instructional coaching partnerships can improve teachers’ implementation of learned mathematics instructional strategies. Teachers are willing to learn and implement new mathematics strategies after professional development sessions to see better student learning results. However, the implementation process can become difficult. Our purpose was to determine whether implementing mathematics strategies improved if an instructional coaching partnership supported teachers.

“Do instructional coaching partnerships improve teachers’ implementation of mathematics instructional strategies?” We gathered data to determine whether instructional coaching partnerships support teachers’ capacity to implement new learning. Data were collected using video recording or classroom observation as a pre- and post-assessment. Teachers received 4 to 6 weeks of instructional coaching support during the intervention. Teachers completed a questionnaire about their intervention experiences. Student testing data were also analyzed to determine whether the intervention increased learning outcomes.

Our findings showed improved mathematics strategies, explicitly implementing the open-ended questioning strategy used during mathematics instruction. Open-ended questions to check students’ mathematics understanding increased by 42%. Teachers responded to a qualitative survey and stated overall satisfaction with the support provided by the instructional coach. Additionally, state testing scores in Grades 3 to 5 increased proficiency levels. Grade-level growth comparisons increased between 5 and 28%.

This study adds to current research stating that instructional coaching cycles and the implementation of partnership principles can positively support the execution of learned teaching practices. The study also indicates the effects of coaching support on students’ learning.