
Data Science for Strategic Advantage Session 2
Udemy ยท English
About this course
- Language
- English
- Category
- Provider
- Udemy
- Data updated
The theme of Session 2 of "Data Science for Strategic Advantage" is hypothesis testing. It is one of the most frequently used statistical methods in practice; for example, setting up a two-alternative hypothesis on whether a product improvement or development is a success or a failure, and making a judgment from a statistical perspective. There are three key points to understanding the principles of hypothesis testing and applying them to real-world problem-solving:โ Formulating action plans and risk countermeasures: Hypothesis testing estimates the frequency of a hypothesis occurring; it does not reveal the "true state" of the hypothesis, nor the "correctness" of the decision. Problem-solving requires action plans and risk countermeasures for the four possible outcomes based on the 2ร2 framework of "True State ร Decision Outcome", and you will master how to formulate them.โก Combining parametric and nonparametric tests: Real-world data from social and natural phenomena do not necessarily follow a normal distribution. For this reason, you need not only parametric tests that assume a normal distribution for the population, but also nonparametric tests that can be applied to skewed distributions or distributions with outliers. You will gain two-tiered analytical skills by combining both approaches.โข Making full use of statistical analysis software: Knowing many methods is meaningless unless you can actually use them. By mastering JASP, a free statistical analysis software developed at the University of Amsterdam, you will acquire practical applied skills through hands-on exercises. This course (Part 1) covers the basic logic of hypothesis testing and one-sample tests, while the next session (Part 2) will cover methods for two or more samples. When you think of data science, various machine learning and deep learning methods likely come to mind, along with the large language models powering today's highly anticipated geโฆ











