klymya/ts_cource_practice

This repository contains practice homework for time-series courses.

This repository contains practice homework for time-series courses.

Practices:

1. Trend analysis

   • Time series smoothing
   • Trend estimation and extraction

2. Trend predictive models(In progress)

   • Naive Approach
   • Simple Exponential Smoothing
   • Holt's Linear Trend Model
   • Holt Winter's Model

3. Dynamic predictive models(In progress)

   • Residual analisys. Dickey-Fuller test
   • Autocorelation analisys
   • Predicting Time series: AR Model
   • Predicting Time series: Moving Average Model

Datasets:

• daily-min-temperatures.csv - the dataset containce the minimum daily temperatures over 10 years (1981-1990) in the city Melbourne, Australia.
• monthly-sunspots.csv - describes a monthly count of the number of observed sunspots for just over 230 years (1749-1983).
• daily-total-female-births.csv - the dataset describes the number of daily female births in California in 1959.
• airline-passengers.csv - the dataset describe number of air passengers per month from 1949 to 1960.
• opsd_germany_daily.csv - contains electricity consumption, wind power production, and solar power production for 2006–2017.

Structure:

• /data - contains csv files with data
• /notebooks - contains - ipynb files with prectices

Requirements:

To complete this practice you need to install Anaconda. Anaconda is a Python data science distribution with preinstalled libraries.

Useful links:

• Machine Learning Mastery. Introduction to Time Series Forecasting (Python)
• Открытый курс машинного обучения. Тема 9. Анализ временных рядов с помощью Python