Added R code for week3 seminar

_freeze/week3/index/execute-results/html.json

@@ -1,9 +1,11 @@
 {
-  "hash": "50901a1a61b70455ef3b4f5e3724c654",
+  "hash": "d1b4fd6e61e389ce60f2a9b600e83f7a",
   "result": {
     "engine": "knitr",
-    "markdown": "---\ntitle: \"Week 3: Time series decomposition\"\n---\n\n::: {.cell}\n\n:::\n\n\n\n\n\n## What you will learn this week\n\n* Transforming data to remove some sources of variation\n* Decomposing a time series into trend-cycle, seasonal and remainder components\n* Seasonal adjustment\n\n## Pre-class activities\n\nRead [Chapter 3 of the textbook](https://otexts.com/fpp3/decomposition.html) and watch all embedded videos\n\n## Exercises (on your own or in tutorial)\n\nComplete Exercises 6-11 from [Section 2.10 of the book](https://otexts.com/fpp3/graphics-exercises.html).\n\n\n\n\n\n\n## Slides for seminar\n\n<iframe src='https://docs.google.com/gview?url=https://af.numbat.space/week3/slides.pdf&embedded=true' width='100%' height=465></iframe>\n<a href=https://af.numbat.space/week3/slides.pdf class='badge badge-small badge-red'>Download pdf</a>\n\n## Seminar activities\n\n\n1.  Produce an STL decomposition as follows\n\n    ```r\n    us_construction <- us_employment |>\n      filter(Title == \"Construction\", year(Month) > 1980)\n    dcmp <- us_construction |>\n      model(stl = STL(Employed ~ trend(window = 9) + season(window = 11)))\n    dcmp |>\n      components() |>\n      autoplot()\n    ```\n\n2. What happens as you change the values of the two `window` arguments?\n\n3. How does the seasonal shape change over time? [Hint: Try plotting the seasonal component using `gg_season`.]\n\n4. Can you produce a plausible seasonally adjusted series? [Hint: `season_adjust` is one of the variables returned by `STL`.]\n\n\n\n## Assignments\n\n* [Assignment 2](../assignments/A2.qmd) is due on Friday 22 March.\n",
-    "supporting": [],
+    "markdown": "---\ntitle: \"Week 3: Time series decomposition\"\n---\n\n::: {.cell}\n\n:::\n\n\n\n\n## What you will learn this week\n\n* Transforming data to remove some sources of variation\n* Decomposing a time series into trend-cycle, seasonal and remainder components\n* Seasonal adjustment\n\n## Pre-class activities\n\nRead [Chapter 3 of the textbook](https://otexts.com/fpp3/decomposition.html) and watch all embedded videos\n\n## Exercises (on your own or in tutorial)\n\nComplete Exercises 6-11 from [Section 2.10 of the book](https://otexts.com/fpp3/graphics-exercises.html).\n\n\n\n\n## Slides for seminar\n\n<iframe src='https://docs.google.com/gview?url=https://af.numbat.space/week3/slides.pdf&embedded=true' width='100%' height=465></iframe>\n<a href=https://af.numbat.space/week3/slides.pdf class='badge badge-small badge-red'>Download pdf</a>\n\n\n\n\n[**R code used in seminar**](seminar_code.R)\n\n## Seminar activities\n\n\n\n\n\n\n1.  Produce an STL decomposition as follows\n\n    ```r\n    us_construction <- us_employment |>\n      filter(Title == \"Construction\", year(Month) > 1980)\n    dcmp <- us_construction |>\n      model(stl = STL(Employed ~ trend(window = 9) + season(window = 11)))\n    dcmp |> components() |> autoplot()\n    ```\n2. What happens as you change the values of the two `window` arguments?\n3. How does the seasonal shape change over time? *[Hint:&nbsp;Try&nbsp;plotting the seasonal component using `gg_season`.]*\n4. Can you produce a plausible seasonally adjusted series? *[Hint:&nbsp;`season_adjust` is returned by `components()`.]*\n\n\n## Assignments\n\n* [Assignment 2](../assignments/A2.qmd) is due on Friday 22 March.\n",
+    "supporting": [
+      "index_files"
+    ],
     "filters": [
       "rmarkdown/pagebreak.lua"
     ],

week3/index.qmd

@@ -21,10 +21,21 @@ Read [Chapter 3 of the textbook](https://otexts.com/fpp3/decomposition.html) and
 
 Complete Exercises 6-11 from [Section 2.10 of the book](https://otexts.com/fpp3/graphics-exercises.html).
 
-
 ```{r}
 #| output: asis
 show_slides(week)
-show_activity(week)
+```
+
+[**R code used in seminar**](seminar_code.R)
+
+## Seminar activities
+
+```{r}
+#| child: activities.md
+```
+
+```{r}
+#| output: asis
+#| echo: false
 show_assignments(week)
 ```

week3/seminar_code.R

@@ -2,10 +2,16 @@ library(fpp3)
 
 # Australian monthly electricity production
 elec1 <- as_tsibble(fma::elec) |>
-  transmute(Month = index, GWh = value)
+  rename(Month = index, GWh = value)
+
+elec1
 
 elec1 |>
-  autoplot(box_cox(GWh, lambda = 0.3))
+  autoplot(GWh)
+
+lambda <- elec1 |> features(GWh, guerrero)
+elec1 |>
+  autoplot(box_cox(GWh, lambda = lambda))
 
 dcmp <- elec1 |>
   model(stl = STL(box_cox(GWh, lambda = 0.3)))
@@ -19,14 +25,16 @@ components(dcmp) |>
 
 elec1 |>
   autoplot(GWh, color = "gray") +
-  autolayer(components(dcmp), inv_box_cox(trend, lambda = 0.3), color = "red")
+  autolayer(components(dcmp), inv_box_cox(trend, lambda = 0.3),
+            color = "red")
 
 elec1 |>
   autoplot(GWh, color = "gray") +
   autolayer(components(dcmp), inv_box_cox(season_adjust, lambda = 0.3), color = "red")
 
 elec1 |>
-  model(STL(log(GWh) ~ season(window = 3333) + trend(window = 7), robust = TRUE)) |>
+  model(STL(log(GWh) ~ season(window = 11) + trend(window = 909995),
+            robust = TRUE)) |>
   components() |>
   autoplot()
 
@@ -68,14 +76,20 @@ us_construction <- us_employment |>
 us_construction |> autoplot(Employed)
 
 dcmp <- us_construction |>
-  model(stl = STL(Employed ~ trend(window = 9) + season(window = 7)))
+  model(stl = STL(Employed ~ trend(window = 9) +
+                    season(window = 9)))
 dcmp |>
   components() |>
   autoplot()
 dcmp |>
   components() |>
   gg_season(season_year)
 
+dcmp |>
+  components() |>
+  as_tsibble() |>
+  autoplot(season_adjust)
+
 
 ## ABS employment problem