'IT Support for Research' Workshop: Machine Learning with Python (2)

A time series is a series of data points listed in time order and frequently appears in various industries including economics, statistics, bioinformatics, etc. The upcoming workshop 'Machine Learning with Python (2)', consists of three lessons, aims at introducing participants the basic concept of time series analysis and forecasting methods with Python.

All students are welcome to join.

'IT Support for Research' Workshop: Machine Learning with Python (2)

Date: 4, 9 & 11 November 2021

Time: 14:30 – 17:00

Target Audience: RPg, TPg and Ug students

Medium of Instruction: English

Course Outline:

Lesson 1 (4 Nov)
- Introduction and Terminology
- Time Series Plot
- Common Time Series Transformation

• Log, Power, Box-Cox Transformations

- Moving Average

• Simple Moving Average and Exponential Moving Average

- Naïve Prediction Method

• Seasonal Naïve Method, Extrapolation

- Time Series Decomposition

- Stationarity

• Augmented Dicky Fuller Test

Lesson 2 (9 Nov)
- Univariate Time Series Forecast

• Exponential Smoothing
  • Single, Double and Triple (Holt’s Winter) Exponential Smoothing
  • Damped Method
• ARIMA Model
  • Autoregression Model, Moving Average Model, Autoregressive Moving Average Model, Non-seasonal and Seasonal ARIMA Model

Lesson 3 (11 Nov)
- Multivariate Time Series Forecast

• Vector Auto Regression (VAR)
  • Order selection
  • Granger’s Causality Test
  • Order of Integration
  • Durbin Watson Statistic
• Long Short-Term Memory (LSTM)
  • Introduction to Neural Network
    • Artificial Neural Network, Deep Neural Network, Recurrent Neural Network
  • Introduction to LSTM
  • Time Series Forecast with LSTM using Keras