Python Code Guide

This appendix provides reusable Python code patterns for NREC4410: International Agricultural Trade. The code is written for teaching, not for advanced software engineering. Copy and adapt the examples when preparing figures, tables, or simple simulations.

The code chunks in this appendix are shown for reference and are not executed during rendering.

Basic imports

Code

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

Create a simple table

Code

trade_data = pd.DataFrame({
    "Country": ["Oman", "India"],
    "Exports": [45.0, 780.0],
    "Imports": [31.0, 720.0]
})

trade_data

Calculate trade balance and openness

Code

exports = 28
imports = 36
gdp = 120

trade_balance = exports - imports
trade_openness = (exports + imports) / gdp * 100

print(f"Trade balance = {trade_balance:.1f} billion USD")
print(f"Trade openness = {trade_openness:.1f}%")

Bar chart for exports and imports

Code

trade_data = pd.DataFrame({
    "Category": ["Agricultural products", "Non-agricultural products"],
    "Exports": [1834, 42718],
    "Imports": [5133, 25747]
})

ax = trade_data.set_index("Category").plot(kind="bar")
ax.set_xlabel("")
ax.set_ylabel("Million US dollars")
ax.set_title("Merchandise trade by category")
plt.xticks(rotation=0)
plt.tight_layout()
plt.show()

Horizontal bar chart for top products

Code

top_exports = pd.DataFrame({
    "Product": [
        "Crude petroleum",
        "Petroleum gases",
        "Refined petroleum",
        "Nitrogenous fertilizers",
        "Motor cars"
    ],
    "Value": [18686, 4403, 3865, 1498, 955]
})

ax = top_exports.plot(kind="barh", x="Product", y="Value", legend=False)
ax.set_xlabel("Million US dollars")
ax.set_ylabel("")
ax.set_title("Top exported products")
ax.invert_yaxis()
plt.tight_layout()
plt.show()

Ricardian PPF

The Ricardian labor constraint is:

\[ a_{LF}Q_F + a_{LC}Q_C = L \]

Solving for cloth:

\[ Q_C = \frac{L - a_{LF}Q_F}{a_{LC}} \]

Code

# Oman example
L = 2000
a_LF = 1
a_LC = 2

food = np.linspace(0, L / a_LF, 100)
cloth = (L - a_LF * food) / a_LC

plt.figure()
plt.plot(food, cloth)
plt.xlabel("Food")
plt.ylabel("Cloth")
plt.title("Production Possibility Frontier")
plt.xlim(0, L / a_LF)
plt.ylim(0, L / a_LC)
plt.tight_layout()
plt.show()

Ricardian PPF and TPF

Code

L = 2000
a_LF = 1
a_LC = 2
world_price_food = 1.0  # 1 food = 1 cloth

food = np.linspace(0, L / a_LF, 100)
ppf_cloth = (L - a_LF * food) / a_LC

# Specialization in food
max_food = L / a_LF
tpf_cloth = world_price_food * (max_food - food)

plt.figure()
plt.plot(food, ppf_cloth, label="PPF")
plt.plot(food, tpf_cloth, linestyle="--", label="TPF")
plt.xlabel("Food")
plt.ylabel("Cloth")
plt.title("PPF and TPF")
plt.legend()
plt.tight_layout()
plt.show()

Consumer surplus and producer surplus

For linear demand and supply:

\[ Q_d = a - bP \]

\[ Q_s = c + dP \]

Code

def demand(P, a, b):
    return a - b * P

def supply(P, c, d):
    return c + d * P

def autarky_equilibrium(a, b, c, d):
    P = (a - c) / (b + d)
    Q = supply(P, c, d)
    return P, Q

def surplus_at_price(P, a, b, c, d):
    Qd = demand(P, a, b)
    Qs = supply(P, c, d)
    P_max = a / b
    P_min = -c / d
    CS = 0.5 * (P_max - P) * Qd
    PS = 0.5 * (P - P_min) * Qs
    TS = CS + PS
    return CS, PS, TS, Qd, Qs

# Example
P, Q = autarky_equilibrium(a=80, b=1, c=0, d=1)
CS, PS, TS, Qd, Qs = surplus_at_price(P, a=80, b=1, c=0, d=1)

print(P, Q, CS, PS, TS)

Plot supply and demand with surplus areas

Code

a, b = 80, 1
c, d = 0, 1
P_eq, Q_eq = autarky_equilibrium(a, b, c, d)

P_values = np.linspace(0, a / b, 200)
Qd_values = demand(P_values, a, b)
Qs_values = supply(P_values, c, d)

plt.figure()
plt.plot(Qd_values, P_values, label="Demand")
plt.plot(Qs_values, P_values, label="Supply")
plt.axhline(P_eq, linestyle="--", label=f"Price = {P_eq:.0f}")
plt.axvline(Q_eq, linestyle="--", label=f"Quantity = {Q_eq:.0f}")
plt.fill_betweenx(P_values, 0, Qd_values, where=P_values >= P_eq, alpha=0.2)
plt.fill_betweenx(P_values, 0, Qs_values, where=P_values <= P_eq, alpha=0.2)
plt.xlabel("Quantity")
plt.ylabel("Price")
plt.title("Consumer and Producer Surplus")
plt.legend()
plt.tight_layout()
plt.show()

Export supply and import demand

Code

# Country 1
params_1 = {"a": 80, "b": 1, "c": 0, "d": 1}

# Country 2
params_2 = {"a": 100, "b": 0.5, "c": 0, "d": 0.5}

P_values = np.linspace(0, 120, 200)

ES = supply(P_values, params_1["c"], params_1["d"]) - demand(P_values, params_1["a"], params_1["b"])
ED = demand(P_values, params_2["a"], params_2["b"]) - supply(P_values, params_2["c"], params_2["d"])

plt.figure()
plt.plot(ES, P_values, label="Export supply")
plt.plot(ED, P_values, label="Import demand")
plt.axhline(60, linestyle="--", label="World price")
plt.axvline(40, linestyle="--", label="Trade volume")
plt.xlabel("Trade quantity")
plt.ylabel("Price")
plt.title("World Equilibrium")
plt.legend()
plt.tight_layout()
plt.show()

Effective rate of protection calculator

Code

def effective_rate_of_protection(output_world, input_world, output_tariff, input_tariffs):
    """
    output_world: world price of final good
    input_world: list of input costs at world prices
    output_tariff: tariff rate on final good, such as 0.20
    input_tariffs: list of tariff rates for each input
    """
    VA_world = output_world - sum(input_world)
    output_domestic = output_world * (1 + output_tariff)
    inputs_domestic = [cost * (1 + tariff) for cost, tariff in zip(input_world, input_tariffs)]
    VA_domestic = output_domestic - sum(inputs_domestic)
    ERP = (VA_domestic - VA_world) / VA_world * 100
    return VA_world, VA_domestic, ERP

# Cheese example
output_world = 20
input_world = [10, 5]
output_tariff = 0.20
input_tariffs = [0.10, 0.10]

effective_rate_of_protection(output_world, input_world, output_tariff, input_tariffs)

FTA welfare calculation

Code

P0 = 6000
P1 = 5500
Q0 = 10  # thousand cars
Q1 = 15  # thousand cars
tariff = 1000

price_drop = P0 - P1
cs_gain = price_drop * Q0 + 0.5 * price_drop * (Q1 - Q0)
tariff_revenue_loss = tariff * Q0
net_welfare = cs_gain - tariff_revenue_loss

print(f"Consumer surplus gain = {cs_gain:.2f} million USD")
print(f"Tariff revenue loss = {tariff_revenue_loss:.2f} million USD")
print(f"Net welfare effect = {net_welfare:.2f} million USD")

Exchange rate and import price

Code

foreign_price_usd = 300
exchange_rate = 0.385  # OMR per USD

domestic_price_omr = foreign_price_usd * exchange_rate
print(f"Domestic price = {domestic_price_omr:.2f} OMR")

Simulated gravity model data

Code

gravity_data = pd.DataFrame({
    "Partner": ["India", "UAE", "Saudi Arabia", "China", "Germany"],
    "GDP_partner": [3900, 500, 1100, 18000, 4200],
    "Distance": [1700, 450, 1200, 5600, 5200],
    "Trade": [5.2, 9.1, 4.8, 6.5, 2.4]
})

gravity_data["ln_trade"] = np.log(gravity_data["Trade"])
gravity_data["ln_gdp_partner"] = np.log(gravity_data["GDP_partner"])
gravity_data["ln_distance"] = np.log(gravity_data["Distance"])

gravity_data

Gravity-style scatter plot

Code

plt.figure()
plt.scatter(gravity_data["Distance"], gravity_data["Trade"])
for _, row in gravity_data.iterrows():
    plt.text(row["Distance"], row["Trade"], row["Partner"])
plt.xlabel("Distance from Oman, km")
plt.ylabel("Trade value")
plt.title("Trade and Distance")
plt.tight_layout()
plt.show()

TINA simulation table

Code

tina_results = pd.DataFrame({
    "Direction": ["Oman exports to India", "India exports to Oman"],
    "Trade creation": [402.80, 550.61],
    "Trade diversion": [164.67, 106.02]
})

tina_results["Total trade effect"] = tina_results["Trade creation"] + tina_results["Trade diversion"]
tina_results

TINA results chart

Code

ax = tina_results.set_index("Direction")[["Trade creation", "Trade diversion"]].plot(kind="bar")
ax.set_ylabel("USD million")
ax.set_title("CEPA simulated trade effects")
plt.xticks(rotation=0)
plt.tight_layout()
plt.show()

Save a figure as SVG

Use SVG when you want a clean figure for the website.

Code

plt.figure()
plt.plot([1, 2, 3], [1, 4, 9])
plt.xlabel("x")
plt.ylabel("y")
plt.title("Example figure")
plt.tight_layout()
plt.savefig("figures/example-figure.svg")
plt.show()

Common Quarto chunk options

#| label: fig-example
#| fig-cap: "Figure caption here."
#| fig-width: 7
#| fig-height: 4
#| echo: true
#| warning: false
#| message: false

Suggested coding rules for this course

Keep examples short.
Use clear variable names.
Label axes and figures.
Use tables before complex graphs.
Avoid unnecessary packages.
Prefer pandas, numpy, and matplotlib.
Do not use code that requires private data.
For final projects, explain the economic interpretation before the code.

--- title: "Python Code Guide" execute: eval: false echo: true warning: false message: false --- This appendix provides reusable Python code patterns for **NREC4410: International Agricultural Trade**. The code is written for teaching, not for advanced software engineering. Copy and adapt the examples when preparing figures, tables, or simple simulations. The code chunks in this appendix are shown for reference and are not executed during rendering. ## Basic imports ```{python} import numpy as np import pandas as pd import matplotlib.pyplot as plt ``` ## Create a simple table ```{python} trade_data = pd.DataFrame({ "Country": ["Oman", "India"], "Exports": [45.0, 780.0], "Imports": [31.0, 720.0] }) trade_data ``` ## Calculate trade balance and openness ```{python} exports = 28 imports = 36 gdp = 120 trade_balance = exports - imports trade_openness = (exports + imports) / gdp * 100 print(f"Trade balance = {trade_balance:.1f} billion USD") print(f"Trade openness = {trade_openness:.1f}%") ``` ## Bar chart for exports and imports ```{python} trade_data = pd.DataFrame({ "Category": ["Agricultural products", "Non-agricultural products"], "Exports": [1834, 42718], "Imports": [5133, 25747] }) ax = trade_data.set_index("Category").plot(kind="bar") ax.set_xlabel("") ax.set_ylabel("Million US dollars") ax.set_title("Merchandise trade by category") plt.xticks(rotation=0) plt.tight_layout() plt.show() ``` ## Horizontal bar chart for top products ```{python} top_exports = pd.DataFrame({ "Product": [ "Crude petroleum", "Petroleum gases", "Refined petroleum", "Nitrogenous fertilizers", "Motor cars" ], "Value": [18686, 4403, 3865, 1498, 955] }) ax = top_exports.plot(kind="barh", x="Product", y="Value", legend=False) ax.set_xlabel("Million US dollars") ax.set_ylabel("") ax.set_title("Top exported products") ax.invert_yaxis() plt.tight_layout() plt.show() ``` ## Ricardian PPF The Ricardian labor constraint is: $$ a_{LF}Q_F + a_{LC}Q_C = L $$ Solving for cloth: $$ Q_C = \frac{L - a_{LF}Q_F}{a_{LC}} $$ ```{python} # Oman example L = 2000 a_LF = 1 a_LC = 2 food = np.linspace(0, L / a_LF, 100) cloth = (L - a_LF * food) / a_LC plt.figure() plt.plot(food, cloth) plt.xlabel("Food") plt.ylabel("Cloth") plt.title("Production Possibility Frontier") plt.xlim(0, L / a_LF) plt.ylim(0, L / a_LC) plt.tight_layout() plt.show() ``` ## Ricardian PPF and TPF ```{python} L = 2000 a_LF = 1 a_LC = 2 world_price_food = 1.0 # 1 food = 1 cloth food = np.linspace(0, L / a_LF, 100) ppf_cloth = (L - a_LF * food) / a_LC # Specialization in food max_food = L / a_LF tpf_cloth = world_price_food * (max_food - food) plt.figure() plt.plot(food, ppf_cloth, label="PPF") plt.plot(food, tpf_cloth, linestyle="--", label="TPF") plt.xlabel("Food") plt.ylabel("Cloth") plt.title("PPF and TPF") plt.legend() plt.tight_layout() plt.show() ``` ## Consumer surplus and producer surplus For linear demand and supply: $$ Q_d = a - bP $$ $$ Q_s = c + dP $$ ```{python} def demand(P, a, b): return a - b * P def supply(P, c, d): return c + d * P def autarky_equilibrium(a, b, c, d): P = (a - c) / (b + d) Q = supply(P, c, d) return P, Q def surplus_at_price(P, a, b, c, d): Qd = demand(P, a, b) Qs = supply(P, c, d) P_max = a / b P_min = -c / d CS = 0.5 * (P_max - P) * Qd PS = 0.5 * (P - P_min) * Qs TS = CS + PS return CS, PS, TS, Qd, Qs # Example P, Q = autarky_equilibrium(a=80, b=1, c=0, d=1) CS, PS, TS, Qd, Qs = surplus_at_price(P, a=80, b=1, c=0, d=1) print(P, Q, CS, PS, TS) ``` ## Plot supply and demand with surplus areas ```{python} a, b = 80, 1 c, d = 0, 1 P_eq, Q_eq = autarky_equilibrium(a, b, c, d) P_values = np.linspace(0, a / b, 200) Qd_values = demand(P_values, a, b) Qs_values = supply(P_values, c, d) plt.figure() plt.plot(Qd_values, P_values, label="Demand") plt.plot(Qs_values, P_values, label="Supply") plt.axhline(P_eq, linestyle="--", label=f"Price = {P_eq:.0f}") plt.axvline(Q_eq, linestyle="--", label=f"Quantity = {Q_eq:.0f}") plt.fill_betweenx(P_values, 0, Qd_values, where=P_values >= P_eq, alpha=0.2) plt.fill_betweenx(P_values, 0, Qs_values, where=P_values <= P_eq, alpha=0.2) plt.xlabel("Quantity") plt.ylabel("Price") plt.title("Consumer and Producer Surplus") plt.legend() plt.tight_layout() plt.show() ``` ## Export supply and import demand ```{python} # Country 1 params_1 = {"a": 80, "b": 1, "c": 0, "d": 1} # Country 2 params_2 = {"a": 100, "b": 0.5, "c": 0, "d": 0.5} P_values = np.linspace(0, 120, 200) ES = supply(P_values, params_1["c"], params_1["d"]) - demand(P_values, params_1["a"], params_1["b"]) ED = demand(P_values, params_2["a"], params_2["b"]) - supply(P_values, params_2["c"], params_2["d"]) plt.figure() plt.plot(ES, P_values, label="Export supply") plt.plot(ED, P_values, label="Import demand") plt.axhline(60, linestyle="--", label="World price") plt.axvline(40, linestyle="--", label="Trade volume") plt.xlabel("Trade quantity") plt.ylabel("Price") plt.title("World Equilibrium") plt.legend() plt.tight_layout() plt.show() ``` ## Effective rate of protection calculator ```{python} def effective_rate_of_protection(output_world, input_world, output_tariff, input_tariffs): """ output_world: world price of final good input_world: list of input costs at world prices output_tariff: tariff rate on final good, such as 0.20 input_tariffs: list of tariff rates for each input """ VA_world = output_world - sum(input_world) output_domestic = output_world * (1 + output_tariff) inputs_domestic = [cost * (1 + tariff) for cost, tariff in zip(input_world, input_tariffs)] VA_domestic = output_domestic - sum(inputs_domestic) ERP = (VA_domestic - VA_world) / VA_world * 100 return VA_world, VA_domestic, ERP # Cheese example output_world = 20 input_world = [10, 5] output_tariff = 0.20 input_tariffs = [0.10, 0.10] effective_rate_of_protection(output_world, input_world, output_tariff, input_tariffs) ``` ## FTA welfare calculation ```{python} P0 = 6000 P1 = 5500 Q0 = 10 # thousand cars Q1 = 15 # thousand cars tariff = 1000 price_drop = P0 - P1 cs_gain = price_drop * Q0 + 0.5 * price_drop * (Q1 - Q0) tariff_revenue_loss = tariff * Q0 net_welfare = cs_gain - tariff_revenue_loss print(f"Consumer surplus gain = {cs_gain:.2f} million USD") print(f"Tariff revenue loss = {tariff_revenue_loss:.2f} million USD") print(f"Net welfare effect = {net_welfare:.2f} million USD") ``` ## Exchange rate and import price ```{python} foreign_price_usd = 300 exchange_rate = 0.385 # OMR per USD domestic_price_omr = foreign_price_usd * exchange_rate print(f"Domestic price = {domestic_price_omr:.2f} OMR") ``` ## Simulated gravity model data ```{python} gravity_data = pd.DataFrame({ "Partner": ["India", "UAE", "Saudi Arabia", "China", "Germany"], "GDP_partner": [3900, 500, 1100, 18000, 4200], "Distance": [1700, 450, 1200, 5600, 5200], "Trade": [5.2, 9.1, 4.8, 6.5, 2.4] }) gravity_data["ln_trade"] = np.log(gravity_data["Trade"]) gravity_data["ln_gdp_partner"] = np.log(gravity_data["GDP_partner"]) gravity_data["ln_distance"] = np.log(gravity_data["Distance"]) gravity_data ``` ## Gravity-style scatter plot ```{python} plt.figure() plt.scatter(gravity_data["Distance"], gravity_data["Trade"]) for _, row in gravity_data.iterrows(): plt.text(row["Distance"], row["Trade"], row["Partner"]) plt.xlabel("Distance from Oman, km") plt.ylabel("Trade value") plt.title("Trade and Distance") plt.tight_layout() plt.show() ``` ## TINA simulation table ```{python} tina_results = pd.DataFrame({ "Direction": ["Oman exports to India", "India exports to Oman"], "Trade creation": [402.80, 550.61], "Trade diversion": [164.67, 106.02] }) tina_results["Total trade effect"] = tina_results["Trade creation"] + tina_results["Trade diversion"] tina_results ``` ## TINA results chart ```{python} ax = tina_results.set_index("Direction")[["Trade creation", "Trade diversion"]].plot(kind="bar") ax.set_ylabel("USD million") ax.set_title("CEPA simulated trade effects") plt.xticks(rotation=0) plt.tight_layout() plt.show() ``` ## Save a figure as SVG Use SVG when you want a clean figure for the website. ```{python} plt.figure() plt.plot([1, 2, 3], [1, 4, 9]) plt.xlabel("x") plt.ylabel("y") plt.title("Example figure") plt.tight_layout() plt.savefig("figures/example-figure.svg") plt.show() ``` ## Common Quarto chunk options ```text #| label: fig-example #| fig-cap: "Figure caption here." #| fig-width: 7 #| fig-height: 4 #| echo: true #| warning: false #| message: false ``` ## Suggested coding rules for this course 1. Keep examples short. 2. Use clear variable names. 3. Label axes and figures. 4. Use tables before complex graphs. 5. Avoid unnecessary packages. 6. Prefer `pandas`, `numpy`, and `matplotlib`. 7. Do not use code that requires private data. 8. For final projects, explain the economic interpretation before the code.