Tomasz Gonciarz

Language: EN Industry: Data analytics Target Audience: data analysts interested in Numpy, Colab and Python How is NumPy used in real-world data analysis? If you’re learning data science with Python, NumPy is one of the most powerful tools at your fingertips. Beyond theory, it shines in practical applications like:  - Cleaning and preprocessing numerical data  - Calculating moving averages and statistical indicators  - Running simulations (e.g. Monte Carlo models)  - Image processing using pixel arrays  - Feature engineering in machine learning workflows  - Fast numerical operations in Exploratory data Analysis (EDA)

2

2 Kommentare

Today I summarized one of the most important concepts in statistical modeling: the difference between parametric and non-parametric regression. Parametric models are simple, fast, and rely on an assumed functional form. Non-parametric models are flexible and allow the data to determine the shape of the relationship. Understanding when to use each approach is essential for building reliable predictive models.

27

2 Kommentare

Exploring 150+ years of Swiss temperature data (1864–2020) to practice time-series analysis tools with Pandas! Using raw historical data from MeteoSwiss, I decided to make a small practice notebook to explore long-term temperature trends in Switzerland to practice time series analysis with real, imperfect data. This was an opportunity to approach a project by cleaning, restructuring, and visualizing. The result: a clear upward trend starting around the 1970s, which accelerates in the last 30 years. You can check the full notebook on GitHub: https://lnkd.in/dPN5SnvH #Python #Pandas #DataAnalysis #ClimateData #TimeSeries #DataScience

4

If you're building a time series forecasting model, choosing the right approach is critical. This post offers a valuable breakdown of 10 top techniques — from statistical models like ARIMA to deep learning architectures like TCN — highlighting their strengths, limitations, and ideal use cases. Thank you, Timur Bikmukhametov, PhD, for sharing these clear and practical insights. #TimeSeries #MachineLearning #Forecasting #DataScience

156

10 Kommentare

Many analytics careers start at university. In this #TalentSpotlight story, Miroslava Zajacová shares how using SAS during her thesis helped her build strong analytical skills – skills she now uses as a Financial Controller. 🚀 Read the full story: https://lnkd.in/eA9yeDW6

5

🚨 The Most Important Step in Data Analysis Isn't What You Think 🚨 It's not building the most complex model. It's not writing the perfect Python script. It's asking the right question first. I've learned that a clear, well-defined business question is the compass for any data project. Without it, you're just exploring a jungle of data without a map. Here's my framework for turning vague requests into analytical projects: 1. From "What's going on?" to → "Which product category showed a sales decline of >10% last quarter?" 2. From "Improve customer satisfaction" to → "What are the top 3 reasons for support tickets, and can we automate solutions for them?" A sharp question leads to a clear hypothesis, which dictates the data you need, the tools you use (hello, #Python & #SQL!), and how you measure success. My question for the data pros: What's your process for scoping a data problem before you write a single line of code? What's the most challenging "vague request" you've had to translate into a data question? Share your wisdom in the comments! 👇 #DataAnalytics #DataScience #BusinessIntelligence #ProblemSolving #DataDriven #MachineLearning #Analytics #CareerAdvice

10

1 Kommentar

🎯 A/B Testing Fundamentals Quiz Think you know your A/B testing basics? From control groups to statistical significance—put your knowledge to the test. ✅ You’ll cover: - Core concepts and terminology - Common pitfalls - Best practices to run effective experiments 👉 𝗧𝗮𝗸𝗲 𝘁𝗵𝗲 𝗾𝘂𝗶𝘇 𝗻𝗼𝘄 >> https://lnkd.in/ePcDCnWy #ABTesting #ProductGrowth #Experimentation

2

1 Kommentar

Choosing the right statistical test shouldn’t feel like guessing. If you’ve ever found yourself asking; “Should I use a t-test or a Mann-Whitney U?” “Is this variable normally distributed?” “Am I working with paired or unpaired data?” “Should i use linear or logistic regression” then this article is for you. “Selection of Appropriate Statistical Methods for Data Analysis” by Dr Prabhaker Mishra et al. is a must read paper especially for students, early career researchers, and anyone trying to make sound statistical decisions in biomedical or public health research. What you’ll learn ✓ When to use parametric vs. non-parametric methods ✓ How to match your research question, data type, and assumptions to the right test ✓ Common pitfalls in statistical interpretation ✓ Why your data's structure (paired/unpaired, distribution, etc.) matters Read the full article here https://lnkd.in/dXvAExWB Remember, good biostatistics starts with appropriate analysis. Tag or share with a friend or colleague who’s always debating between t-test and Wilcoxon. #biostatistics #medicalresearch #dataanalysis #researchtips #parametricVsnonparametric #publichealth #statisticalliteracy #quantitativeresearch #Stata #RStats #researchmethods

116

4 Kommentare