Tanat Tonguthaisri, CISSP®’s Post

Discover how complexity in analytical science can be both a challenge and an opportunity. Should we fight it or embrace it? Dive into the intricacies and learn how to transform complexity into clarity. Read the blog article by Camilla Liscio - https://ow.ly/vgzz50SK9nU Join us at Future Focus III on September 25 & 26 to explore this topic further! More info and tickets - https://ow.ly/Qf7A50SK9nV #AnalyticalScience #DataAnalytics #Complexity #Innovation #FutureFocusIII #SamplePreparation #SampleAnalysis #ElementLabSolution

Data-driven decisions are both science and art. Join us and RXA @ OneMagnify for a free webinar on 1/17 for a discussion on how to strike the balance. Register here: https://okt.to/nPg3EH #datadriven

Navigate Market Volatility with Precision! 🌟   Is volatile data causing your predictions to turn out flat? Dive into our latest #whitepaper, where we introduce a revolutionary ensemble model. By merging the strengths of multiple time series models, the new ensemble model significantly enhances both predictive #accuracy and #reliability, while maintaining model explainability.   Uncover how this innovative strategy enables institutions to master complex market dynamics and make decisions with heightened #precision. Ideal for #financialanalysts, #datascientists, and anyone involved in strategic decision-making, our whitepaper provides the tools to revolutionize your #forecasting techniques.   🔗 Explore our insights: https://lnkd.in/g3FfzEUN   #AdvancedForecasting #DataScience #MachineLearning #SARIMA #LSTM #StrategicInsights #BusinessIntelligence #InnovativeSolutions

Check out this insightful blog post on conditional entropy, which reveals its almost time-reversal invariance and its impact on quantifying learnability and controlling for distributional shift in datasets. The note provides valuable guidelines for practitioners. Read the full post here: https://bit.ly/49oano5

NEW QUADRANT REPORT: Analytical Accelerators and Tools, 2024 This report, part of Chartis’ STORM research series, explores the evolving landscape of analytical accelerators and tools designed to mitigate computational bottlenecks and enhance the performance of financial models and analytics. Check it out here: https://lnkd.in/gRuHG2ZV #AnalyticalAccelerators #Analytics #FinancialModels #FinancialRisk #Research

Easily identify changes in causal structures over time and discover causal graphs for each time window 🫡✅ Causal structures evolve over time in most real-world problems Domain experts/Business stakeholders often find it unrealistic to fit 1 causal graph in the entire dataset & lose their trust in the methods We recently added the just-in-time causal discovery framework to address these challenges Inspired by the paper “Causal Discovery for Non-stationary Non-linear Time-series Data Using Just-In-Time Modeling” (link: https://lnkd.in/e4n2EQgi), users can now combine the just-in-time framework with 10+ causal discovery algos to identify the time points at which the causal structure changes + discover the causal graph in each window Great work by our R&D team and a useful feature for applying causal discovery in the real world. Try it out in our free trial: https://lnkd.in/e4PUjkcA

Designing randomized trials might be challenging. Why sequential trials are especially challenging. "People’s lives are not cross sectional, they are a trajectory through time. So there are lots of design decisions to be made in order to squeeze all of it into tabular form. Having longitudinal data makes it more possible, but not more easy. How do we decide how to assign a person into either treatment groups (or not at all if not eligible)." writes Ehud Karavani (IBM Research) in his new blog post on sequential trial design for causal inference. His blog is a great high-level introduction to the topic. It also comes with a comprehensive list of resources. See 👇🏼 for details.

Life sciences organizations are quickly adopting digital initiatives to accelerate innovation. To successfully transform, organizations need better data management and to break down data silos while protecting intellectual property and ensuring compliance. In the first part of Iron Mountain's Life Sciences Webinar series, it outlines how to adopt a “digital-first” mindset for greater success in 2024: http://spr.ly/6044phpKu #digitaltransformation #innovation #datamanagement #lifesciences

🚀 New Blog: Navigating High Cardinality in Metrics 🚀 High cardinality data offers precision but can inflate storage and computational costs. Our latest blog dives into how to manage this balance effectively in observability metrics. Key tips include: 🔹 Avoid Improper Instrumentation 🔹 Optimize Histogram Buckets 🔹 Use Consistent Tags ❇ Check out Harold Lim's insightful blog here: https://lnkd.in/ekFhrftu #observability #o11y #metrics #cardinality And stay tuned for our upcoming blog, where we’ll tackle high cardinality in events, logs, and traces!