Roll-over for 2025

Author: robjhyndman

_dependencies.R

@@ -2,3 +2,4 @@ library(commonmark)
 library(markdown)
 library(languageserver)
 library(codetools)
+library(moodlequiz)

_quarto.yml

@@ -32,10 +32,10 @@ website:
           - week12/index.qmd
       - text: Discussion
         icon: chat-fill
-        href: https://edstem.org/au/courses/15863/discussion/
+        href: https://edstem.org/au/courses/21163/discussion/
       - text: Moodle
         icon: mortarboard-fill
-        href: https://learning.monash.edu/course/view.php?id=10091&section=1
+        href: https://learning.monash.edu/course/view.php?id=26695&section=1
       - text: Textbook
         icon: book-fill
         href: https://OTexts.com/fpp3

assignments.csv

@@ -1,6 +1,6 @@
 Assignment, Due, Moodle, File
-Assignment 1, 2024-03-08, 2298116, "A1.qmd"
-Assignment 2, 2024-03-24, 2034165, "A2.qmd"
-Assignment 3, 2024-04-14, 2034169, "A3.qmd"
-Assignment 4, 2024-05-05, 2034170, "A4.qmd"
-Retail Project, 2024-05-24, 2034167, "Project.qmd"
+Forecasting Competition, 2025-03-14, 3444036, "competition.qmd"
+Assignment 1, 2025-03-28, 3444030, "A1.qmd"
+Assignment 2, 2025-04-18, 3444031, "A2.qmd"
+Assignment 3, 2025-05-16, 3444032, "A3.qmd"
+Retail Project, 2025-05-30, 3444033, "Project.qmd"

assignments/A1.qmd

@@ -1,57 +1,37 @@
 ---
 title: Assignment 1
-fig-cap-location: top
 ---
 
-**You must provide forecasts for the following items:**
+This assignment will use the same data that you will use in the [retail project](Project.qmd) later in semester. Each student will use a different time series, selected using their student ID number as follows.
+
+```r
+library(fpp3)
+get_my_data <- function(student_id) {
+  set.seed(student_id)
+  all_data <- readr::read_rds("https://bit.ly/monashretaildata")
+  while(TRUE) {
+    retail <- filter(all_data, `Series ID` == sample(`Series ID`, 1))
+    if(!any(is.na(fill_gaps(retail)$Turnover))) return(retail)
+  }
+}
+# Replace the argument with your student ID
+retail <- get_my_data(12345678)
+```
+
+  1. Plot your time series using the `autoplot()` command. What do you learn from the plot?
+  2. Plot your time series using the `gg_season()` command. What do you learn from the plot?
+  3. Plot your time series using the `gg_subseries()` command. What do you learn from the plot?
+  4. Find an appropriate Box-Cox transformation for your data and explain why you have chosen the particular transformation parameter $\lambda$.
+  5. Produce a plot of an STL decomposition of the transformed data. What do you learn from the plot?
 
-  1. Google closing stock price on 20 March 2024 [[Data](https://finance.yahoo.com/quote/GOOG/)].
-  2. Maximum temperature at Melbourne airport on 10 April 2024 [[Data](http://www.bom.gov.au/climate/dwo/IDCJDW3049.latest.shtml)].
-  3. The difference in points (Collingwood minus Essendon) scored in the AFL match between Collingwood and Essendon for the Anzac Day clash. 25 April 2024 [[Data](https://en.wikipedia.org/wiki/Anzac_Day_match)].
-  4. The seasonally adjusted estimate of total employment for April 2024. ABS CAT 6202, to be released around mid May 2024 [[Data](https://www.abs.gov.au/statistics/labour/employment-and-unemployment/labour-force-australia/latest-release)].
-  5. Google closing stock price on 22 May 2024 [[Data](https://finance.yahoo.com/quote/GOOG/)].
+For all plots, please use appropriate axis labels and titles.
 
-**For each of these, give a point forecast and an 80% prediction interval, and explain in a couple of sentences how each was obtained.**
+You need to submit one Quarto (`qmd`) file which implements all steps above.
 
-* The [Data] links give you possible data to start with, but you are free to use any data you like.
-* There is no need to use any fancy models or sophisticated methods. Simple is better for this assignment. The methods you use should be understandable to any high school student.
-* Full marks will be awarded if you submit the required information, and are able to meaningfully justify your results in a couple of sentences in each case.
-* Once the true values in each case are available, we will come back to this exercise and see who did the best using the scoring method described in class.
-* The student with the lowest score is the winner of our forecasting competition, and will win a $50 cash prize.
-* The assignment mark is not dependent on your score.
+To receive full marks, the `qmd` file must compile without errors.
 
 ```{r}
 #| output: asis
 source(here::here("course_info.R"))
 submit(schedule, "Assignment 1")
 ```
-
-___
-
-## Leaderboard
-
-```{r}
-readRDS(here::here("assignments/A1_leaderboard.rds")) |>
-  DT::datatable()
-```
-
-## Forecasts
-
-::: {layout-ncol=5}
-
-### Q1
-[![Google closing stock price on 20 March 2024](Q1.png)](Q1.png)
-
-### Q2
-[![Maximum temperature at Melbourne airport on 10 April 2024](Q2.png)](Q2.png)
-
-### Q3
-[![Difference in points in AFL Anzac Day clash](Q3.png)](Q3.png)
-
-### Q4
-[![Seasonally adjusted total employment for April 2024](Q4.png)](Q4.png)
-
-### Q5
-[![Google closing stock price on 22 May 2024](Q5.png)](Q5.png)
-
-:::

assignments/A1_leaderboard.rds

@@ -10,25 +10,22 @@ get_my_data <- function(student_id) {
   set.seed(student_id)
   all_data <- readr::read_rds("https://bit.ly/monashretaildata")
   while(TRUE) {
-    retail <- filter(all_data, `Series ID` == sample(`Series ID`, 1)) 
+    retail <- filter(all_data, `Series ID` == sample(`Series ID`, 1))
     if(!any(is.na(fill_gaps(retail)$Turnover))) return(retail)
   }
 }
 # Replace the argument with your student ID
 retail <- get_my_data(12345678)
 ```
 
-  1. Plot your time series using the `autoplot()` command. What do you learn from the plot?
-  2. Plot your time series using the `gg_season()` command. What do you learn from the plot?
-  3. Plot your time series using the `gg_subseries()` command. What do you learn from the plot?
-  4. Find an appropriate Box-Cox transformation for your data and explain why you have chosen the particular transformation parameter $\lambda$.
-  5. Produce a plot of an STL decomposition of the transformed data. What do you learn from the plot?
+1. Using a test set of 2019--2022, fit an ETS model chosen automatically, and three benchmark methods to the training data. Which gives the best forecasts on the test set, based on RMSE?
+2. Check the residuals from the best model using an ACF plot and a Ljung-Box test. Do the residuals appear to be white noise?
+3. Now use time-series cross-validation with a minimum sample size of 15 years, a step size of 1 year, and a forecast horizon of 5 years. Calculate the RMSE of the results. Does it change the conclusion you reach based on the test set?
+4. Which of these two methods of evaluating accuracy is more reliable? Why?
 
-For all plots, please use appropriate axis labels and titles.
+Submit a Quarto (`qmd`) file which carries out the above analysis. You need to submit one file which implements all steps above.
 
-You need to submit one Rmarkdown or Quarto file which implements all steps above.
-
-To receive full marks, the Rmd or qmd file must compile without errors.
+To receive full marks, the `qmd` file must compile without errors.
 
 ```{r}
 #| output: asis

assignments/A1_responses.rds

@@ -2,26 +2,32 @@
 title: Assignment 3
 ---
 
-This assignment will use national population data from 1960 -- 2022. Each student will use a different time series, selected using their student ID number as follows.
+This assignment will use the same data that you will use in the [retail project](Project.qmd) later in semester. Each student will use a different time series, selected using their student ID number as follows.
 
 ```r
 library(fpp3)
-# Replace seed with your student ID
-set.seed(12345678)
-pop <- readr::read_rds("https://bit.ly/monashpopulationdata") |>
-  filter(Country == sample(Country, 1))
+get_my_data <- function(student_id) {
+  set.seed(student_id)
+  all_data <- readr::read_rds("https://bit.ly/monashretaildata")
+  while(TRUE) {
+    retail <- filter(all_data, `Series ID` == sample(`Series ID`, 1))
+    if(!any(is.na(fill_gaps(retail)$Turnover))) return(retail)
+  }
+}
+# Replace the argument with your student ID
+retail <- get_my_data(12345678)
 ```
 
-Population should be modelled as a logarithm as it increases exponentially.
+Use a training set up to and including 2018.
 
-1. Using a test set of 2018--2022, fit an ETS model chosen automatically, and three benchmark methods to the training data. Which gives the best forecasts on the test set, based on RMSE?
-2. Check the residuals from the best model using an ACF plot and a Ljung-Box test. Do the residuals appear to be white noise?
-3. Now use time-series cross-validation with a minimum sample size of 15 years, a step size of 1 year, and a forecast horizon of 5 years. Calculate the RMSE of the results. Does it change the conclusion you reach based on the test set?
-4. Which of these two methods of evaluating accuracy is more reliable? Why?
+* What transformations (Box-Cox and/or differencing) would be required to make the data stationary? You should use a unit-root test as part of the discussion.
+* Use a plot of the ACF and PACF of the (possibly differenced) data to determine two plausible models for this data set.
+* Fit both models, along with an automatically chosen model, and produce forecasts for 2019--2022.
+* Which model is best based on AIC? Which model is best based on the test set RMSE? Which do you think is best to use for future forecasts? Why?
+* Check the residuals from your preferred model, using an ACF plot and a Ljung-Box test. Do the residuals appear to be white noise?
 
-Submit an Rmd or qmd file which carries out the above analysis. You need to submit one file which implements all steps above.
+Submit a Quarto (`qmd`) file which carries out the above analysis. You need to submit one file which implements all steps above.
 
-To receive full marks, the Rmd or qmd file must compile without errors.
 
 ```{r}
 #| output: asis