Infopaket

Statistical data analysis

ВБ 1.07

Вибіркова дисципліна для ОП

First

2021/2022

7 Semester

4

The student must know the basic operators and approaches of the Python language, data types, data visualization; working with the stats module, statsmodels and scipy packages; PANDAS module tools; AR autoregression model; equations for OMV, properties of estimates; autoregression model with residuals in the form of a moving average; formulas for coefficients; testing of hypotheses about the order of AR, MA, ARIMA, Foster-Stewart test. The student must be able to program in the Python language, be able to write code for the implementation of basic statistical methods of data analysis and their visualization; apply in practice the construction of a histogram, basic and chain rates growth; apply the Foster-Stewart criterion, the series criterion; to investigate linear and polynomial models; choose the parameters of the ARMA process using the AKAIKE criterion; construct ACF, PACF, apply the Dickey-Foulier test.

Full-time form

The educational discipline "Statistical data analysis" is based on the cycle of disciplines "Theory of Probability", "Applied Mathematical Statistics", "Statistical Radiophysics", "Object Oriented Programming". Prerequisites: the student should know: the basics of probability theory and mathematical statistics, basics programming. the student must be able to: operate with random variables and processes, apply statistical methods in practice, programming.

The discipline "Statistical data analysis" allows the student to orient himself in modern problems of applied statistics and learn the software system that is a tool for solving them tasks As you know, statistical methods are the most widely used mathematical tool for specialists in the field of economics. The course "Statistical data analysis" consists of sections "Basic concepts of the Python language", "Autoregressive models", "Moving average models", "SARIMA models", "Markov chains", "Diffusion processes". Each of these sections allows you to learn certain statistical methods of analysis data both from a theoretical point of view and from a practical point of view, based on the software environment (languages programming) Python. It is a very powerful and popular programming language that is spreading free in the world and rapidly developed by users

1. Майборода Р.Є. «Комп’ютерна статистика.» Підручник. - К., ВПЦ «Київський університет», 2019. https://probability.knu.ua/userfiles/mre/cscolor.pdf 2. Майборода Р.Є., Сугакова О.В. «Аналіз даних за допомогою пакету R». Навчальний посібник – К., 2015. http://matphys.rpd.univ.kiev.ua/wp/wp-content/uploads/2016/12/Statistics_with_R.pdf 3. Т. Андерсон. «Статистика часовіх рядов» – Boca Raton, London, New York: CRC Press, Taylor&Francis Group, 2008. – 700 p. 4. Черняк О.І., Комашко О.В., Ставицький А.В., Баженова О.В. «Економетрика» - ВПЦ «Київський університет», 2009. 5. Василик О.І., Яковенко Т.О. «Лекції з теорії і методів вибіркових обстежень» - К., ВПЦ «Київський університет», 2010 – 208 с. 6. ARIMA models https://pyflux.readthedocs.io/en/latest/arima.html

Lectures, aboratory works, independent work.

Semester assessment: an academic semester has one meaningful module. Student must complete and pass four laboratory works. Mandatory for admission to the exam is: score at least 36 points during the semester. Final evaluation in the form of credit, the form of credit is written and oral. The condition for achieving a positive evaluation is discipline is to receive at least 60 points, the exam grade cannot be less than 24 points.

Ukrainian