DS803: Statistics for Data Science

Students taking the course are recommended  to:

• Have knowledge of mathematics at high school level
• Be able to use PC

The course cannot be chosen by students who: Have passed courses in basic or advanced statistics courses at the level of ST520 Applied statistics or higher.

• Understand concepts in probability and distribution theory.
• Utilize graphics and summary methods for descriptive data analysis.
• Describe data using key statistics such as mean, variance, and correlation.
• Construct confidence intervals for key statistics.
• Test simple statistical hypotheses.
• Analyze data using simple regression models.
• Design data collection.
• Understand central elements in published results from statistical analyses of actual data.
• Critically evaluate the appropriateness of employed methods and inferences based on these.
• Present statistical results in non-technical terms.
• Use the statistical software R for analysing actual data, which is important in regard to being able to work academically with data science problems.

• Give the competence to working critically with own projects and data.
• Give skills to critically evaluate scientific publications.
• Give knowledge and understanding of choice and use of appropriate statistical methods.

• Utilizing graphics and summary methods for descriptive data analysis.
• Describing data using key statistics such as mean, variance, and correlation.
• Constructing confidence intervals for key statistics.
• Testing simple statistical hypotheses.
• Analyzing data using simple regression models.
• Designing data collection.
• Understanding central elements in published results from statistical analyses of actual data.
• Critically evaluating the appropriateness of employed methods and inferences based on these.
• Presenting statistical results in non-technical terms.
• Use R for simple statistical analyses.

• The foundation for statistical considerations. 
• From population to sample and back again. 
• Basic parameters and their estimation. 
• Descriptive statistics (tables and graphics). 
• Basal calculus 
• Probabilities and distributions. 
• Hypotheses and principles for tests. 
• Examples of test methods: t-test, chi-square-test. 
• Basic concepts underlying linear models starting from simple linear regression. 
• Basic concepts with regard to study design. 
• Common problems in applied statistics (types of inferential error, mass significance).
• In the course the statistical software R is used.

• Intro phase (lectures) - 26 hours
• Training phase: 22 hours, including 11 hours tutorials and 11hours laboratory

• Work on specific problems not covered in the training phase hours.
• Discussion of the terms and concepts and problems in regard to data collection and data quality.