rmd file of wheat exercice

module3/exo3/wheat.Rmd

+---
+title: "wheat"
+author: "Fouad Gasmi"
+date: "2025-02-18"
+output: html_document
+---
+
+
+```{r }
+# --------------------------------------------------------------------
+# Setup chunk for R Markdown
+# --------------------------------------------------------------------
+knitr::opts_chunk$set(echo = TRUE, warning = FALSE, message = FALSE)
+```
+
+```{r packages}
+# --------------------------------------------------------------------
+# Check if packages are installed; if not, install them
+# --------------------------------------------------------------------
+required_packages <- c("ggplot2", "dplyr")
+for(p in required_packages){
+  if(!requireNamespace(p, quietly = TRUE)){
+    install.packages(p)
+  }
+}
+
+library(ggplot2)
+library(dplyr)
+```
+
+```{r load data}
+# --------------------------------------------------------------------
+# Load data
+# --------------------------------------------------------------------
+df <- read.csv("data/Wheat.csv", header = TRUE)
+
+# Inspect structure
+str(df)
+```
+
+```{r adding data of monarch reign}
+# --------------------------------------------------------------------
+# Create a data frame of monarch reigns (for background shading)
+# --------------------------------------------------------------------
+monarch_df <- data.frame(
+  name = c("Elizabeth", "James I", "Charles I", "Cromwell",
+           "Charles II", "James II", "W&M", "Anne",
+           "George I", "George II", "George III", "George IV"),
+  start = c(1565, 1603, 1625, 1649,
+            1660, 1685, 1689, 1702,
+            1714, 1727, 1760, 1820),
+  end   = c(1603, 1625, 1649, 1660,
+            1685, 1689, 1702, 1714,
+            1727, 1760, 1820, 1821)
+)
+
+monarch_df$fill_color <- ifelse(
+  seq_len(nrow(monarch_df)) %% 2 == 1, 
+  "white", 
+  "gray90"
+)
+```
+
+```{r Plot theme}
+# --------------------------------------------------------------------
+# Minimalistic theme
+# --------------------------------------------------------------------
+theme_playfair <- theme_minimal(base_size = 12) + 
+  theme(
+    panel.grid.major = element_line(color = "grey80"),
+    panel.grid.minor = element_line(color = "grey90")
+  )
+
+```
+
+```{r First graph reproducing, fig.width=12, fig.height=5}
+# --------------------------------------------------------------------
+# First Plot (Bars for Wheat, Area & Line for Wages)
+# --------------------------------------------------------------------
+g1 <- ggplot(df, aes(x = Year)) +
+  # Bars for Wheat prices
+  geom_bar(
+    aes(y = Wheat), 
+    stat = "identity", 
+    fill = "grey70", 
+    width = 4,
+    alpha = 0.8
+  ) +
+  # Area for Wages
+  geom_area(aes(y = Wages), fill = "lightblue", alpha = 0.6) +
+  # Red line on top of the area for Wages
+  geom_line(aes(y = Wages), color = "red", linewidth = 1) +
+  labs(
+    x = "Year",
+    y = "Shillings (Combined Wheat & Wages)",
+    title = "Graph Replicating William Playfair's Idea",
+    subtitle = "Wheat price (bars) and wages (area/line) on the same scale"
+  ) +
+  theme_playfair
+
+# Display first version
+print(g1)
+
+# Add monarch backgrounds and labels
+g1 <- g1 +
+  geom_rect(
+    data = monarch_df,
+    aes(
+      xmin = start,
+      xmax = end,
+      ymin = -Inf,
+      ymax = Inf,
+      fill = fill_color
+    ),
+    inherit.aes = FALSE,
+    alpha = 0.5,
+    color = NA
+  ) +
+  scale_fill_identity() +
+  geom_text(
+    data = monarch_df,
+    aes(
+      x = (start + end)/2,
+      y = Inf,
+      label = name
+    ),
+    inherit.aes = FALSE,
+    vjust = 1.2,
+    size = 3,
+    color = "black",
+    fontface = "bold"
+  )
+
+# Display updated plot
+print(g1)
+```
+
+```{r second graph using two axis, fig.width=12, fig.height=5}
+# --------------------------------------------------------------------
+# Second Plot with Two Axes (Wheat vs. Wages)
+# --------------------------------------------------------------------
+
+# Calculate scaling factor (to align Wages visually to Wheat scale for the same y-axis range)
+scale_factor <- max(df$Wheat, na.rm = TRUE) / max(df$Wages, na.rm = TRUE)
+
+g2 <- ggplot(df, aes(x = Year)) +
+  geom_line(aes(y = Wheat), color = "blue", size = 1) +
+  geom_line(aes(y = Wages * scale_factor), color = "red", size = 1) +
+  scale_y_continuous(
+    name = "Wheat Price (shillings per quarter bushel)",
+    sec.axis = sec_axis(
+      trans = ~ . / scale_factor,
+      name = "Weekly Wages (shillings per week)"
+    )
+  ) +
+  labs(
+    x = "Year",
+    title = "Wheat Price and Wages on Two Axis",
+    subtitle = "Left Axis: shillings/quarter bushel, Right Axis: shillings/week"
+  ) +
+  theme_playfair
+
+print(g2)
+
+# Add monarch backgrounds and labels
+g2 <- g2 +
+  geom_rect(
+    data = monarch_df,
+    aes(xmin = start, xmax = end, ymin = -Inf, ymax = Inf, fill = fill_color),
+    alpha = 0.5,
+    inherit.aes = FALSE
+  ) +
+  scale_fill_identity() +
+  geom_text(
+    data = monarch_df,
+    aes(x = (start + end)/2, y = Inf, label = name),
+    inherit.aes = FALSE,
+    vjust = 1.2,
+    size = 3
+  )
+
+print(g2)
+```
+
+```{r other representation of second plot, fig.width=12, fig.height=5}
+# --------------------------------------------------------------------
+# Third Plot: Bars for Wheat, Line for Wages on Two Axes
+# --------------------------------------------------------------------
+
+scale_factor <- max(df$Wheat, na.rm = TRUE) / max(df$Wages, na.rm = TRUE)
+
+g2b <- ggplot(df, aes(x = Year)) +
+  # Bars for Wheat
+  geom_bar(
+    aes(y = Wheat),
+    stat = "identity",
+    width = 4,
+    fill = "grey70",
+    alpha = 0.8
+  ) +
+  # Red line for Wages (scaled)
+  geom_line(
+    aes(y = Wages * scale_factor),
+    color = "red",
+    size = 1
+  ) +
+  scale_y_continuous(
+    name = "Wheat Price (shillings per quarter bushel)",
+    sec.axis = sec_axis(
+      trans = ~ . / scale_factor,
+      name = "Weekly Wages (shillings per week)"
+    )
+  ) +
+  labs(
+    x = "Year",
+    title = "Wheat Price (bars) and Wages (line) on Two Axes",
+    subtitle = "Main Axis: Wheat, Secondary Axis: Wages"
+  ) +
+  theme_playfair
+
+print(g2b)
+
+# Add monarch backgrounds and labels
+g2b <- g2b +
+  geom_rect(
+    data = monarch_df,
+    aes(
+      xmin = start,
+      xmax = end,
+      ymin = -Inf,
+      ymax = Inf,
+      fill = fill_color
+    ),
+    inherit.aes = FALSE,
+    alpha = 0.5
+  ) +
+  scale_fill_identity() +
+  geom_text(
+    data = monarch_df,
+    aes(
+      x = (start + end)/2,
+      y = Inf,
+      label = name
+    ),
+    inherit.aes = FALSE,
+    vjust = 1.2,
+    size = 3,
+    color = "black"
+  )
+
+print(g2b)
+```
+
+```{r third plot purchasing power, fig.width=12, fig.height=5}
+# --------------------------------------------------------------------
+# Fourth Plot: Purchasing Power = Wages / Wheat
+# --------------------------------------------------------------------
+
+df$PurchasingPower <- df$Wages / df$Wheat
+
+g3 <- ggplot(df, aes(x = Year, y = PurchasingPower)) +
+  geom_line(color = "darkgreen", size = 1) +
+  geom_point(color = "darkgreen", size = 2) +
+  labs(
+    x = "Year",
+    y = "Quarters of Bushels of Wheat per Weekly Wage",
+    title = "Evolution of Workers' Purchasing Power (in Wheat Volume)",
+    subtitle = "Inspired by Playfair's demonstration of rising purchasing power over time"
+  ) +
+  theme_playfair
+
+print(g3)
+
+# Add monarch backgrounds and labels
+g3 <- g3 +
+  geom_rect(
+    data = monarch_df,
+    aes(xmin = start, xmax = end, ymin = -Inf, ymax = Inf, fill = fill_color),
+    alpha = 0.5,
+    inherit.aes = FALSE
+  ) +
+  scale_fill_identity() +
+  geom_text(
+    data = monarch_df,
+    aes(x = (start + end)/2, y = Inf, label = name),
+    inherit.aes = FALSE,
+    vjust = 1.2,
+    size = 3
+  )
+
+print(g3)
+```
+
+```{r forth plot wheat vs wages without time axis, fig.width=12, fig.height=5}
+# --------------------------------------------------------------------
+# Fifth Plot: Scatter/Path of Wheat vs. Wages (Time as a Color Gradient)
+# --------------------------------------------------------------------
+
+# Make sure df is sorted by Year
+df <- df[order(df$Year), ]
+
+g4 <- ggplot(df, aes(x = Wheat, y = Wages)) +
+  geom_path(
+    aes(color = Year),
+    arrow = arrow(type = "open", length = unit(0.15, "inches")),
+    size = 1
+  ) +
+  geom_point(aes(color = Year), size = 2) +
+  # Color gradient from oldest (blue) to newest (red)
+  scale_color_gradient(low = "blue", high = "red") +
+  labs(
+    x = "Wheat Price (shillings/quarter bushel)",
+    y = "Weekly Wages (shillings/week)",
+    color = "Year",
+    title = "Relationship Between Wheat Price and Weekly Wages (No Direct Time Axis)",
+    subtitle = "Color and arrow indicate chronological progression"
+  ) +
+  theme_minimal(base_size = 12)
+
+print(g4)
+```
+
+```{r , fig.width=12, fig.height=5}
+# --------------------------------------------------------------------
+# Summarize by Decade, Then Plot (Path + Arrow + Labels)
+# --------------------------------------------------------------------
+df_decade <- df %>%
+  mutate(decade = floor(Year / 10) * 10) %>%
+  group_by(decade) %>%
+  summarize(
+    Wheat = mean(Wheat, na.rm = TRUE),
+    Wages = mean(Wages, na.rm = TRUE)
+  ) %>%
+  ungroup()
+
+g_better <- ggplot(df_decade, aes(x = Wheat, y = Wages)) +
+  geom_path(
+    arrow = arrow(length = unit(0.15, "inches"), type = "open"),
+    color = "darkblue",
+    size = 1
+  ) +
+  geom_point(color = "darkblue", size = 3) +
+  geom_text(
+    aes(label = decade),
+    hjust = -0.1,
+    vjust = -0.5,
+    color = "black",
+    size = 3
+  ) +
+  labs(
+    x = "Wheat Price (shillings/quarter bushel)",
+    y = "Weekly Wages (shillings/week)",
+    title = "Wheat Price vs. Wages (Aggregated by Decade)",
+    subtitle = "Arrows and labels indicate progression over time (no direct time axis)"
+  ) +
+  theme_minimal(base_size = 12)
+
+print(g_better)
+```
+
+I find g3 does a better job of illustrating purchasing power. g4 and g_better become too cluttered between 1560 and 1700, making it difficult to interpret the data at a glance. Even when the data is aggregated by decade (g4 vs. g_better), it remains challenging to grasp the information quickly.