Statistics II

Schedule MIM UW:

• 1-2 III, Descriptive statistics - one variable, Exploration
• 8-9 III, Descriptive statistics - two variables, Verification, Exploration, Lab
• 15-16 III [Exploration], Grammar of graphics Theory, Tools: ggplot2, Lab (there will be no classes - 16 III),
• 22-23 III Project presentation at CNK, [Project 1 phase 1]
• 5-6 IV [Exploration], Cluster analysis, k-means, PAM, agnes
• 12-13 IV [Exploration], Multidimensional scaling, [Project 1 phase 2]
• 19-20 IV [Exploration], Decision rules
• 26-27 IV [Prediction], Regression, kNN, trees, model quality measures, [Project 1 phase 3]
• 10-11 V [Prediction], Regression with many variables, model selection,
• 17-18 V [Prediction], Regularisation, feature extraction, [Project 2 phase 1]
• 24-25 V [Prediction], Classification LDA, QDA, logistic regression
• 31-1 V [Prediction], SVM [Project 2 phase 2]
• 7-8 VI [Prediction], Bagging, boosting, random forest
• 14-15 VI [Verification], Multiple hypotheses testing, [Project 2 phase 3]

Projects:

Both projects should be executed in teams. Each team should have 3 or 4 people. You should find different teams for the second projects.

Projects should be presented during each phase. Projects that are not presented will not be scored.

• Project 1 -

Exploration and segmentation for data from CNK Science Center.

Phase 1: Create a single page overview (format A3) for selected stations (one, few or all, it's up to you). The overview should present distribution of playing times with different machines. Consider identification of variables that may influence this variable (day of week, hour, month). Bring the printed overview to the presentation. We will discuss different solutions and approaches.

• Project 2 -

Exploration and prediction for epidemiological data.

Grade:

Each project will be graded on a scale 0-60 points. From projects you can get a pass (maximum grade 4).

Projects are performed in groups but different people would get different number of points, depending on their contribution.

For higher grade you need to pass a written exam.

Data sets:

Will be used during classes.

poslowie <- archivist::aread("pbiecek/Przewodnik/arepo/07088eb35cc2c9d2a2a856a36b3253ad")

Small Logs: http://bit.ly/1X7A2X0