Study unit

Show courses/exams
3622352 Probabilistic inference for data science 1, 5 Cp 
Code 3622352  Validity 01.01.1950 -
Name Probabilistic inference for data science 1  Abbreviation DTN1 
Credits5 Cp  Date of expiry  
TypeIntermediate Studies Subject0390 Statistics 
ClassStudy Unit  Hours  
Study right   Grading0-5 
Recommended scheduling 
   
Organisation Computer Science (J,K) 

Description
Learning outcomes 

Goal of the course is to learn basics of probability calculus for data science, especially mathematical formulation of univariate and multivariate random variables and the required matrix calculus.

 
Content 

The course covers the essential probability calculus for data science. Univariate distributions, transformations and their distribution, expected value, variance and standard deviation. Bivariate distributions, conditional distributions, marginal distributions, covariance, correlation and independence. Matrix calculus. Multivariate distribution, its expected value, variance-covariance matrix and conditional and marginal distributions. Multivariate normal distribution and its essential properties

 
Modes of study 

Self-study based on web material (115 h), lecture (2 h), contact teaching 14 h, written examination 4 h.

 
Teaching methods 

Student prepares for the weekly tutorials by (1) reading given material and watching videos at moodle, (2) doing a related weekly exam and (3) starting the weekly exercises. In tutorials there is also possibility to make some questions on weekly topics. The student (4) returns the weekly exercise report to moodle by the deadline and (5) after the submission deadline, reads the correct answers and self-evaluated his/her own report.  Overall, the students need approximately 15 hours per week for the course, in addition to the 2 hour contact sessions.

 
Study materials 

Handout and other material available in Moodle. Casella-Berger 2002, Statistical inference (chapters 1-4) or DeGroot and Schervish, Probability and Statistics (Chapters 1-4)

 
Evaluation criteria 

Grade (0-5) is based either (1) on a weighted mean of weekly exercises, weekly exams and written exam or (2) on the written exam only, so that the grade is the better one out of these two.

 
Prerequisites 

Introduction to statistics, R-course

 
Time 

Fall semester

 
Offering data 

Course is open to anyone. No restrictions in the number of participants. Useful also for PhD students who need statistical methods in their research

 
Further information 

Teaching language: English

 


Letter (J, K) in front of the name of the course/exam indicates the campus on which teaching or exam takes place: J = Joensuu, K = Kuopio.

Present and future teaching
Functions Name Type Cp Teacher Timetable
Registration not started (KUOPIO): Probabilistic inference for data science 1  Multi-modal teaching  Mika Hujo  26.10.20 -29.01.21
Registration not started (JOENSUU): Probabilistic inference for data science 1  Multi-modal teaching  Mika Hujo  26.10.20 -29.01.21

Future exams
Functions Name Type Cp Teacher Timetable
Register (JOENSUU): Probabilistic inference for data science 1  General examination  Mika Hujo 
20.08.20thu 12.00-16.00
Register (KUOPIO): Probabilistic inference for data science 1  General examination  Mika Hujo 
20.08.20thu 12.00-16.00