Stat 674: Time Series

Department of Statistics and Biostatistics, CSU East Bay

Spring 2023:

Finals Week:

While the due dates for everything are posted, if you need until then of the weekend that is ok. Please have your final versions of everything submitted in Canvas by Sunday March 12.

Week 7:

Hello everyone, I am still not feeling 100% and now that I am starting to feel better I do not want to get sick again. With the cold rain today I think I should teach one more day online before returning to the classroom. I will be teaching on Zoom today and will have office hours on Zoom also (the link is in Canvas as an Announcement). I will not be in the classroom today during class. Please find a place to access Zoom.
Homework: Homework 6 has been posted.
Final: The Final will be made available on Wednesday that will be due at the end of next week. (If you downloaded the .zip before Wednesday, please see the Assignments webpage because the problems for the Final have been changed.)
This week we will discuss ARIMA modeling.
From Rob Hyndman’s Github he has shared Chapter 9 ARIMA Models

Week 6:

One more day online. I am still feeling sick today.
Hello everyone, I woke up sick again on Friday and I am still not feeling well. I will be teaching on Zoom today and will have office hours on Zoom also (the link is in Canvas as an Announcement). I will not be in the classroom today during class. Please find a place to access Zoom.
Homework: Homework 5 has been posted.
Dowload the Author’s slides: Author’s slides We will be using his slides for Regression, Exponential Smoothing, and ARIMA.
Quarto Notebook: Examples of stock market time serise data.
- tidyquant_ver01.qmd

Week 5:

Quiz: There will be a take-home quiz this week that will be due at the end of the week.
Homework: If you have forgotten to turn in a homework for the class, please submit your homework late for consideration. We will go over homework solutions on Wednesday.
Project: The Project has been posted. See the Homework link.
Today we will continue the discussion about Forecasting models. Today we will discuss Residual Diagnostics and Prediction Intervals.
Quarto Notebook:
- Models_ver02.html
- Models_ver02.qmd
Quiz: Download and unzip the Quiz R Project. Complete the quiz in the Quarto Notebook. Render the Quarto Notebook to a .pdf or .docx file. Submit the .qmd file and either the .pdf or .docx file on Canvas. Due Monday Feb. 20, 2023 in Canvas.
- Stat674_Quiz01.zip
Midterm: The Midterm will be made available on Wednesday that will be due at the end of next week.
Homework: Homework 4 has been posted.
Today we will go over the Chapter 4 homework related to simple statistics and return to the discussion of forecasting from Chapter 5. On Wednesday will go over the Quiz and the Midterm.
Midterm:
- Stat674_Midterm.zip
Twitter Thread ACF: Allison Horst Github Allison Horst Twitter ACF
Dowload the Author’s slides: Author’s slides We will be using his slides for Regression, Exponential Smoothing, and ARIMA.

Week 4:

One more day online. I am still feeling sick today.
Hello everyone, I woke up with a head cold this morning. I will be teaching on Zoom today and will have office hours on Zoom also (the link is in Canvas as an Announcement). I will not be in the classroom today during class. Please find a place to access Zoom.
Homework: Homework 3 has been posted.
Today we will discuss Time Series Features. In particular the strength of trend F_t and the strength of seasonality F_s.
Quarto Notebook:
- features_ver01.html
- features_ver01.qmd
Look at the first Forecasting models we will be using in the class.
Quarto Notebook:
- Models_ver01.html
- Models_ver01.qmd

Week 3:

Homework: Homework 2 has been posted.
Today we will start with questions about Homework 1. We will discuss the Classical Decomposition and Seasonal Adjustment.
US Census:
- Seasonal Adjustment Questions and Answers
- Time Series and Seasonal Adjustment
R package:
- seasonalview
- seasonalview.website
Today we will discuss Time Series Components resulting from the Decompositions of a time series.

Week 2:

Book: fpp3
Today we will begin discussing Forecasting and Time Series Analysis. We will begin with some examples and review/introduce the basics of the Tidyverse that are useful for working with tsibble and the other tidyverts R packages. Start by installing the fpp3 R package.
Examples of time series data.
Spotlight book: r4ds
We will be examining many of the datasets we will be using in the class. Note that most of the datasets contain more than one time series.
Quarto Notebook:
- Tsibble_Examples.html
- Tsibble_Examples.qmd
Examples of time series data. What happened to GameStop Corporation’s stock price during the lock-down. Question: How do we load stock data into R? This is a good question.
Time Series plots that show seasonality.
Quarto Notebook:
Spotlight Wikipedia:
- Correlation Coefficient
- Autocorrelation
On Wednesday we will discuss Autocorrelation, White Noise, and the Moving Average Model MA(1).
Quarto Notebook:
- acf_ver01.html
- acf_ver01.qmd

Week 1:

Classroom: NSc119 and Zoom, see Canvas for the link.
Book: fpp3
Reference Book: You can access it through the CSUEB Library. Shumway and Stoffer, Time Series: A Data Analysis Approach Using R, CRC Press, 2019.
Today we will be discussing the syllabus for the class and the material that will be covered.
Homework: Homework 1 has been posted.
Spotlight Software:
- R
- RStudio
Spotlight Blog post: Prime Hints For Running A Data Project In R