# STATISTICAL SCIENCES FOR FINANCE | ADVANCED STATISTICAL LEARNING I

## STATISTICAL SCIENCES FOR FINANCE ADVANCED STATISTICAL LEARNING I

 0222400035 DEPARTMENT OF ECONOMICS AND STATISTICS EQF7 STATISTICAL SCIENCES FOR FINANCE 2022/2023

 OBBLIGATORIO YEAR OF COURSE 1 YEAR OF DIDACTIC SYSTEM 2014 AUTUMN SEMESTER
SSD CFU HOURS ACTIVITY TYPE OF ACTIVITY SECS-S/01 5 30 LESSONS SUPPLEMENTARY COMPULSORY SUBJECTS
 PIETRO CORETTO T
ExamDate
ADVANCED STATISTICAL LEARNING I05/04/2023 - 09:30
ADVANCED STATISTICAL LEARNING I05/04/2023 - 09:30
Objectives
KNOWLEDGE AND UNDERSTANDING

THE COURSE INTRODUCES ADVANCED AUTOMATED STATISTICAL LEARNING METHODS (MACHINE LEARNING) WITH AN EMPHASIS TO APPLICATIONS FOR MASSIVE MODERN DATA SETS (BIG DATA) IN THE FIELD OF ECONOMICS AND FINANCE. AFTER COMPLETING THE COURSE THE STUDENT WILL BE ABLE TO DEVELOP AUTOMATIC DATA ANALYSIS PROCEDURES TO EXTRACT PATTERNS, TRENDS AND FORMULATE PREDICTIONS BASED ON THE USE OF HIGH-DIMENSIONAL AND COMPLEX DATABASES. IN PARTICULAR IT IS EXPECTED THAT A STUDENT ACQUIRE:
-KNOWLEDGE OF MODERN ALGORITHMS FOR SUPERVISED LEARNING
-KNOWLEDGE OF MODERN ALGORITHMS FOR UNSUPERVISED LEARNING
-KNOWLEDGE OF MOST MODERN AND POPULAR SOFTWARE LIBRARIES
-ABILITY TO VALIDATE AND INTERPRET SOLUTIONS PROVIDED BY AN ALGORITHM IN THE SPECIFIC CONTEXT

APPLYING KNOWLEDGE AND UNDERSTANDING

ON THE BASIS OF ACQUIRED KNOWLEDGE THE STUDENT WILL BE ABLE TO
-UNDERSTAND THE TECHNICAL ASPECT UNDERLYING THE MAIN ALGORITHMS FOR STATISTICAL LEARNING
-SELECT AND APPLY THE APPROPRIATE TOOL DEPENDING ON THE CONTEXT AND SPECIFIC PROBLEM TO BE SOLVED
-ABILITY TO USE THE MAIN COMPUTATIONAL TOOLS FOR SOLVING RELEVANT PRACTICAL PROBLEMS OFTEN ARISING IN THE FIELD OF FINANCE AND ECONOMICS
Prerequisites
-MAIN TOOLS IN MATHEMATICAL ANALYSIS
-MATRICES AND MATRIX ALGEBRA
-PROBABILITY
-DESCRIPTIVE STATISTICS AND DATA VISUALIZATION
Contents
THE COURSE WILL COVER THE FOLLOWING TOPICS

- RECALLS ON ASYMPTOTIC THEORY
- INTEGRAL APPROXIMATIONS VIA MONTE CARLO METHODS
- SIMULATION AND COMPUTER-BASED EXPERIMENTS
- TREES-BASED METHODS
- BOOSTING ED ADABOOST
- RANDOM FOREST AND ENSEMBLE LEARNING
- PARTITION METHODS AND SOFT-CLUSTERING
- PARTION METHODS AND SOFT-CLUSTERING ALGORITHMS
- FINITE MIXTURE MODELS
- MODEL-BASED CLUSTERING
Teaching Methods
LECTURES, LAB CLASSES AND CASE STUDIES
Verification of learning
THE FINAL EXAM CONSISTS OF A WRITTEN AND AN ORAL EXAM. BOTH PARTS WILL BE EVALUATED ON A NUMERICAL SCALE BETWEEN 1 AND 30. TO ACCESS THE ORAL PART A MINIMUM OF 18/30 IS REQUIRED FOR THE WRITTEN PART. DURING THE WRITTEN TEST (WHICH LAST ABOUT 2H) THE STUDENT WILL RECEIVE AN EXAM PAPER AND WILL BE ASKED TO ANSWER ABOUT 8 QUESTIONS (EACH WITH SCORES RANGING FROM 1 POINT TO 7 POINTS) ON THE ENTIRE PROGRAM OF THE COURSE, USING A DATASET PROVIDED DURING THE EXAM. THE ORAL EXAM (LASTING ABOUT 15 MINUTES) FOCUSES ON THE GENERAL KNOWLEDGE OF THE TOPICS TREATED DURING THE COURSE, THE ABILITY TO PRODUCE A CORRECT STATISTICAL ANALYSIS, THE ABILITY TO CORRECTLY COMMUNICATE THE RESULTS. THE FINAL MARK WILL REFLECT THE EFFECTIVINESS OF THE TOOLS EMPLOYED, OF THE THOROUGHNESS AND LUCIDITY OF ANSWERS. THE FINAL MARK, ON A SCALE BETWEEN 1 AND 30 WITH LAUDE, WILL CONSIDER BOTH THE PERFORMANCE ON THE WRITTEN AND ORAL PART.
Texts
LECTURES NOTES OF THE INSTRUCTOR AND REFERENCES AVAILABLE ON-LINE SUGGESTED BY THE INSTRUCTOR

FRIEDMAN, J., HASTIE, T., AND TIBSHIRANI, R. (2001). THE ELEMENTS OF STATISTICAL LEARNING (VOL. 1, NO. 10). NEW YORK: SPRINGER SERIES IN STATISTICS.