Short-Run-AD-AS-Model

This model was developed for the short-run macroeconomic analysis in my Intermediate Macroeconomics course.

---

Economics-Notebook

Files

• shinyapp_ch13_AD_AS.R
  This is the main file for the Chapter 13 model.

• ch13_master.qmd
  This file contains older models as well as current and additional code for Chapter 13 and/or the Short-Run model.

---

App Structure

1. App Launch / Global Setup

1. Packages and Scripts Loaded
   • Libraries are loaded and source("modules/simulation_calc.R") is executed.
2. Helper Function Defined
   • initializeExperimentSet(...) is defined.
3. UI Defined
   • The UI is laid out using page_sidebar(...).
   • The main panel includes several tab panels: World, Plots, Log/Ratio Scale, and Program Results (with subtabs).
     At this point, the UI is rendered in the browser, but no simulation data has been generated yet.

---

2. Server Function Initialization

When s(ui, server) starts, the server function is invoked once. Inside the server:

1. Excel Import
   • excel_list <- import_list("data/savings_rate_y.xlsx") loads available Excel sheets for the “World” tab.
   • Observers for regions, years, and the button to calculate the average savings rate are initialized.

---

3. Experiments Table Management

• Create a reactiveVal to store experiments.
• Observe the “Add” button to insert a new row.
• Render the experiment table with “Delete” buttons.
• Observe “Delete” events to remove rows.

---

4. Simulation Triggers (Event Reactive Blocks)

When the user clicks Simulate:

1. simulate_first_exp_calculations
   • An eventReactive(input$simulate, { ... }) block is triggered.
   • It checks if the first experiment set is empty. If empty, a modal error is shown; otherwise, it calls Module Call 1 (simulate_solow) to run the simulation with the parameters and experiment set from Tab 1.
2. simulate_counter_exp_calculations
   • Runs Module Call 1 again, but with no experiments.
3. simulate_second_exp_calculations, simulate_third_exp_calculations, simulate_fourth_exp_calculations
   • Each block uses eventReactive(input$simulate, { ... }) to run the same Module Call 1 but with experiments from sets 2, 3, or 4 (if available; otherwise, it returns NULL).

All these eventReactive blocks return their respective data frames of results when the user clicks Simulate.

---

5. Output Tables (Reactive)

1. Tables Rendered
   • Output slots such as results_first_exp_df, results_counterfactual_df, results_second_exp_df, results_third_exp_df, and results_fourth_exp_df use renderTable({ req(...); ... }).
   • Each table displays the data frame returned by its respective simulation.

---

6. Plot Generation and Display

1. make_plot(...) Function
   • A local helper function that constructs a ggplot2 plot.
2. plot_objects ReactiveValues
   • A reactiveValues() object is used to store each ggplot output for inclusion in the ZIP.
3. plot_specs List
   • A list of plot definitions (IDs, y-variable names, titles, labels).
4. Loop to Render Plots
   • A for (spec in plot_specs) loop sets up output$plot_... <- renderPlot({ ... }).
   • Each time the user clicks Simulate or toggles the checkboxes, these plots refresh automatically if any reactive dependencies change.

---

7. Download Handlers

1. Download Data (downloadData)
   • When clicked, this handler pulls the first experiment’s data from simulate_first_exp_calculations() and writes it to a CSV file.
2. Download All Plots (downloadPlots)
   • This handler allows the user to download a ZIP file containing all the plots as PNGs.

---

8. User Interaction Flow Summary

A typical user flow might look like this:

1. The user launches the app, and the UI appears.
2. The user selects a region/country in the World tab; observers recalculate the valid years.
3. The user clicks Get Avg Savings Rate; the average is computed and populates the slider.
4. The user enters experiments in any of the 4 tabs (Sets 1, 2, 3, 4), either by manually adding rows or by uploading a file.
5. The user clicks Simulate, which:
   • Triggers all the eventReactive blocks:
     • The first scenario is simulated; if there are no rows in Set 1, an error is shown.
     • The counterfactual scenario is simulated (with no experiments).
     • Additional scenarios for sets 2–4 (if rows exist) are simulated.
   • All result data frames (results_*_df) refresh.
   • All plots refresh via renderPlot.
6. The user views the new results in the Program Results tab and sees updated plots in the Plots and Log/Ratio Scale tabs.
7. Optionally, the user toggles checkboxes to show or hide lines for the second, third, and fourth scenarios (or the counterfactual), and the plots update automatically.
8. Optionally, the user clicks Download Results to obtain a CSV of the first scenario or Download All Plots to download a ZIP file of PNGs.