Анализ временных рядов на Python

Looking to master time series analysis? ITtensive offers a specialized course to enhance your machine learning skills.

To gain access to this and other ITtensive courses on Udemy, please email [email protected] with the name(s) of the course(s) you're interested in.

This course delves into practical applications of time series forecasting, covering three hands-on projects:

• Grain Futures Price Prediction: Forecast grain prices using historical data from the London Exchange. We'll apply classic methods like moving averages and polynomial regression to predict prices during periods of high uncertainty. Project: June 2022 Grain Futures Price Forecast.
• Currency Exchange Rate Forecasting: Explore frequency and econometric approaches to predict the USD to RUB exchange rate. Learn to decompose time series into trend, seasonality, and variation. We'll utilize models like ARMA, ARIMA, SARIMA, and vector data, and experiment with libraries like Prophet and Auto-TS (for automated machine learning). Project: December 2022 Export Volume Forecast.
• Electricity Consumption Prediction: Utilize recurrent neural networks to forecast electricity consumption patterns based on a relatively stationary time series. Project: Stock Price Prediction using Recurrent Neural Networks.

Course Theory:

The course covers essential concepts and techniques in time series analysis, including:

• Understanding the fundamentals and goals of time series analysis.
• Basic techniques: Polynomial trends and moving averages.
• The Holt-Winters model and colored noise.
• Autoregressive (AR), Moving Average (MA), ARIMA, SARIMA, ADL, and VAR models.
• Time series analysis methodology and data drift.
• Recurrent Neural Networks (RNNs): LSTM, GRU, ConvLSTM, and BiLSTM.
• An overview of advanced models like WaveNet and Transformers (attention mechanisms).

This course provides a practical, in-depth understanding of time series forecasting techniques, ideal for data scientists and machine learning engineers looking to apply these skills to real-world problems. Improve your predictive modeling capabilities and gain valuable insights from sequential data.