us wedding costs by state data from https://www.markbroumand.com/pages/research-wedding-cost-and-marriage-length interestng paper diamonds are forever that goes into more individual data https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2501480

data-FdN4y.csv

weddingusa

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt

df = pd.read_csv("/content/data-FdN4y.csv")

plot_df = df.rename(columns={
    "Average Wedding Cost": "wedding_cost_usd",
    "Average Marriage Length": "marriage_length_years",
    "State": "state"
}).copy()

x = plot_df["wedding_cost_usd"].to_numpy()
y = plot_df["marriage_length_years"].to_numpy()

corr = np.corrcoef(x, y)[0, 1]
m, b = np.polyfit(x, y, 1)

label_states = set()
label_states.update(plot_df.nlargest(3, "wedding_cost_usd")["state"].tolist())
label_states.update(plot_df.nsmallest(3, "wedding_cost_usd")["state"].tolist())
label_states.update(plot_df.nlargest(3, "marriage_length_years")["state"].tolist())
label_states.update(plot_df.nsmallest(3, "marriage_length_years")["state"].tolist())
if "DC" in plot_df["state"].values:
    label_states.add("DC")

fig, ax = plt.subplots(figsize=(10, 6))

ax.scatter(x, y)

# Trend line
xx = np.linspace(x.min(), x.max(), 200)
ax.plot(xx, m*xx + b)

ax.set_xlabel("Average wedding cost $")
ax.set_ylabel("Average marriage length in years")
ax.set_title("Wedding cost vs. marriage length (by US state)")

# Correlation as a small in-plot note (top-left)
ax.text(
    0.02, 0.98, f"Correlation r = {corr:.2f}",
    transform=ax.transAxes,
    va="top", ha="left",
    fontsize=9
)

# Footer/source line (bottom-left, outside axes a touch)
fig.text(
    0.01, 0.01,
    "data: markbroumand graph: @iamreddave",
    ha="left", va="bottom",
    fontsize=9
)

# Labels for selected points
for _, row in plot_df.iterrows():
    st = row["state"]
    if st in label_states:
        ax.annotate(
            st,
            (row["wedding_cost_usd"], row["marriage_length_years"]),
            textcoords="offset points",
            xytext=(8, 6),
            ha="left",
            fontsize=9
        )

plt.tight_layout(rect=[0, 0.03, 1, 1])  # leave room for footer
plt.show()