Understanding data and statistics correctly: The basics of exploratory data analysis
.svg)
.svg)
.svg)
Machine Learning Engineer
1. sharpen business understanding and questions
- Data-driven decisions versus gut feeling
- Procedure: Key business question → Hypothesis → Metric
- Examples: Churn reduction, conversion boost etc.
2. data profiling and descriptive statistics
- Quality criteria: Data quality, cardinality, completeness
- Position and dispersion key figures: Mean value, median, standard deviation
- Skewness: Recognize and exclude outliers
- Visual explorative data analysis with histogram and boxplot
3. bivariate analysis and visualization in detail
- Reading Pearson correlation and Spearman correlation correctly
- Visualize data with scatter plot, heatmap and grouped bars
- Watch out for pitfalls: Anscombe Quartet and Simpson's Paradox
4. basics: probability and confidence
- Population, sample and random error
- Interpreting confidence intervals
- Discrete and continuous distributions in everyday life
5. hypothesis tests and A/B testing
- Null hypothesis, p-value and significance in plain text
- Comparisons of means and frequencies (Chi² test) without formulas
- Exercise: Evaluate and discuss newsletter experiment
6. outlook on further topics
- Linear regression
- Logistic regression
- Multivariate analysis
- Inductive statistics
Understanding statistics properly: You will learn the basics of descriptive statistics and close the gaps in your knowledge for your data analysis and data science projects.
Make statistics explainable: You will translate formulas into everyday language and learn how to make medians, p-values etc. understandable to non-experts.
Reliable basis for decision-making: You recognize data traps and quantify uncertainty with confidence intervals.
Immediately applicable workflows: From the question to the visualization, you master every step and can, for example, accompany A/B tests independently.
More convincing analyses and visualizations: Sound statistical knowledge makes your evaluations and reports more convincing and facilitates collaboration with data experts.
This training training is conducted in a group of a maximum of 12 participants using the Zoom video conferencing software.
The topics of the course are learned in small exercise projects. Either Microsoft Excel or Google Sheets will be used.
Your trainers are themselves experienced data analysts and data consultants - in the training you will receive concrete, applicable knowledge with many case studies.
You will have space for your questions - individual support from the speaker is guaranteed.
You can access further materials in your personal learning environment.
analysts, product and marketing managers, prospective data scientists and experienced data professionals without formal statistical training who want to interpret figures with confidence and make data-based decisions. No prior knowledge is required.
.svg)
