HRDF Course- Time Series Forecasting in R

Time series are all around us, from server logs, marketing and sales to high-frequency financial data. Time series analysis involves analyzing the available data to o quantitatively monitor, describe, explain, and find out the pattern or trend in the data to predict some future values. Business forecasts can be based on historical data patterns that are used to understand the past and predict future market behavior, inclusing seasonal patterns, trend estimations and growth.

In this course, learn the core techniques necessary to extract meaningful insights from time series data to make informed decision, making. The instructor will cover the basics of time series, analysis techniques, modelling, visualization and forecasting of time series.

Module 1: Basics of Time Series

• Why study time series?
• What is time series modelling?
• Applications of time series modelling
• Time Series vs machine learning

Module 2: Formatting Time Series

• TS stats and queries
• Subsetting and merging
• Decompose and aggregation
• Lags and differences
• Rolling window

Module 3: Smoothing and Forecasting

• Smoothing methods
• Time series models
• Machine learning models
• Examining forecast errors

Module 4: Visualizations

• Basic Charting
• ACF/PACF plots
• Lag plots

Module 5; Abnormality

• Anomality Plots
• Adjusting Decomposition Trend and Seasonality
• Parameter Adjustment
• Adjusting Alpha and Max Anoms
• Detecting anomalities in remainders

Module 6: Sequence Analysis

• Rendering Sequences
• Transition States
• Sequence complexity measures
• Cluster Analysis
• Forecasting pattern sequence