mtahir19/time-series-analysis-03

ARIMA Modeling

---

Materials We Provide

Topic	Description	Link
Lesson	Jupyter Notebook	Link

Dataset description: Global Mean Temperature Data (1750-2015)

---

Learning Objectives

After this lesson, students will be able to:

1. Describe the purpose of the autoregressive and moving average components.
2. Define hyperparameters p, d, and q.
3. Describe AIC.
4. Find the right value of p and q using AIC.
5. Find the right value of d using the augmented Dickey-Fuller test.
6. Complete a manual GridSearch.
7. Fit an ARIMA model.

---

Student Requirements

Before this lesson(s), students should already be able to:

1. Define time series data.
2. Construct autocorrelation and partial autocorrelation plots.
3. Describe autocorrelation and partial autocorrelation.

---

Lesson Outline

Total Time: 120 mins

---

OPTIONAL: Resources for Practice and Learning

For supplemental reading material on this topic, check out the following resources:

• Open-source textbook that's relatively easy to read
• What is second order differencing?
• NIST Documentation on Time Series Analysis
• statsmodels Documentation for ARIMA Model
• statsmodels Documentation for SARIMAX Model
• Wikipedia Article for Augmented Dickey-Fuller Test
• Guidelines for Tuning Parameters p, d, q
• Paper on Finding Parameters of Moving Average Models
• Table for Identifying Which Model and Method of Differencing to Use
• Peter Craigmile's Course on Time Series Analysis

---