AI for financial analysis and forecasting in AI for Everyone - Mini Project: Build & Apply

AI for Financial Analysis and Forecasting

📖 Scenario: You work in a finance team that wants to use AI to help analyze past sales data and predict future sales. This will help the company plan better and make smarter decisions.

🎯 Goal: Build a simple step-by-step understanding of how AI can be used for financial analysis and forecasting by organizing data, setting parameters, applying AI logic, and completing the forecasting setup.

📋 What You'll Learn

Create a data structure with monthly sales figures

Add a threshold value for minimum sales to consider

Use AI logic to filter and prepare data for forecasting

Complete the setup by defining the forecast period

💡 Why This Matters

🌍 Real World

Companies use AI to analyze past financial data and predict future trends to make better business decisions.

💼 Career

Financial analysts and data scientists use these skills to prepare reports and forecasts that guide company strategy.

1

DATA SETUP: Create monthly sales data

Create a dictionary called monthly_sales with these exact entries: 'January': 12000, 'February': 15000, 'March': 13000, 'April': 17000, 'May': 16000.

AI for Everyone

# Create the monthly_sales dictionary with exact values
# Your code here

Hint

Use curly braces to create a dictionary and separate each month and sales value with a colon.

2

CONFIGURATION: Set minimum sales threshold

Create a variable called min_sales_threshold and set it to 14000 to filter out months with low sales.

AI for Everyone

monthly_sales = {
    'January': 12000,
    'February': 15000,
    'March': 13000,
    'April': 17000,
    'May': 16000
}
# Set the minimum sales threshold
# Your code here

Hint

Assign the number 14000 to the variable named min_sales_threshold.

3

CORE LOGIC: Filter months meeting the sales threshold

Create a dictionary called filtered_sales that includes only months from monthly_sales where the sales are greater than or equal to min_sales_threshold. Use a dictionary comprehension with month and sales as variables.

AI for Everyone

monthly_sales = {
    'January': 12000,
    'February': 15000,
    'March': 13000,
    'April': 17000,
    'May': 16000
}
min_sales_threshold = 14000
# Filter sales using dictionary comprehension
# Your code here

Hint

Use a dictionary comprehension with for month, sales in monthly_sales.items() and an if condition.

4

COMPLETION: Define forecast period

Create a variable called forecast_months and set it to 3 to represent the number of months to forecast ahead.

AI for Everyone

monthly_sales = {
    'January': 12000,
    'February': 15000,
    'March': 13000,
    'April': 17000,
    'May': 16000
}
min_sales_threshold = 14000
filtered_sales = {month: sales for month, sales in monthly_sales.items() if sales >= min_sales_threshold}
# Set the forecast period
# Your code here

Hint

Assign the number 3 to the variable named forecast_months.

Practice

1. What is the primary role of AI in financial analysis?

easy

A. To analyze data and predict future financial trends

B. To replace all human financial advisors

C. To create new financial regulations

D. To manually enter financial data

Solution

1. Step 1: Understand AI's function in finance

   AI processes large amounts of financial data to find patterns.

2. Step 2: Identify AI's main benefit

   It helps predict future trends, aiding decision-making.

3. Final Answer:

   To analyze data and predict future financial trends -> Option A

4. Quick Check:

   AI predicts trends = To analyze data and predict future financial trends [OK]

Hint: AI predicts trends by analyzing data patterns [OK]

Common Mistakes:

• Thinking AI replaces all humans
• Confusing AI with regulation creation
• Believing AI only inputs data manually

2. Which of the following is a correct example of AI use in financial forecasting?

easy

A. Using AI to predict stock prices based on historical data

B. Using AI to print physical money

C. Using AI to manually count cash

D. Using AI to write financial laws

Solution

1. Step 1: Identify valid AI applications in finance

   AI analyzes data to forecast trends like stock prices.

2. Step 2: Eliminate incorrect options

   Printing money, manual counting, and law writing are not AI tasks.

3. Final Answer:

   Using AI to predict stock prices based on historical data -> Option A

4. Quick Check:

   AI forecasts stocks = Using AI to predict stock prices based on historical data [OK]

Hint: AI forecasts by analyzing past data, not physical tasks [OK]

Common Mistakes:

• Confusing AI with physical or manual tasks
• Assuming AI creates laws
• Ignoring data analysis role

3. Consider this scenario: An AI model predicts sales will increase by 10% next quarter based on past trends. What does this prediction imply?

medium

A. Sales will definitely increase by exactly 10%

B. Sales might increase, but the prediction is based on data patterns and not guaranteed

C. Sales will decrease because AI always predicts the opposite

D. Sales data is irrelevant to AI predictions

Solution

1. Step 1: Understand AI prediction nature

   AI uses past data to estimate future trends but cannot guarantee exact outcomes.

2. Step 2: Interpret the prediction

   The 10% increase is a likely scenario, not a certainty.

3. Final Answer:

   Sales might increase, but the prediction is based on data patterns and not guaranteed -> Option B

4. Quick Check:

   AI predictions estimate, not guarantee [OK]

Hint: AI predictions are estimates, not certainties [OK]

Common Mistakes:

• Assuming AI predictions are always exact
• Believing AI predicts opposite outcomes
• Ignoring data relevance

4. An AI system for fraud detection flagged many transactions as fraudulent, but most were legitimate. What is the likely issue?

medium

A. The AI system is not connected to the internet

B. The AI is perfect and all flagged transactions are fraudulent

C. The AI model has a high false positive rate and needs better training data

D. The AI model is ignoring all data

Solution

1. Step 1: Analyze the problem with flagged transactions

   Many legitimate transactions flagged means false positives are high.

2. Step 2: Identify cause and fix

   Improving training data quality can reduce false positives.

3. Final Answer:

   The AI model has a high false positive rate and needs better training data -> Option C

4. Quick Check:

   High false positives = need better training [OK]

Hint: Too many false alerts mean training data needs improvement [OK]

Common Mistakes:

• Assuming AI is always perfect
• Blaming internet connection
• Thinking AI ignores data

5. A financial company wants to use AI to forecast quarterly revenue but has incomplete and inconsistent data. What should they do to improve AI forecasting accuracy?

hard

A. Use AI immediately without checking data quality

B. Delete all old data and start fresh without any records

C. Ignore AI and rely only on manual calculations

D. Clean and organize the data, then combine AI predictions with expert human insights

Solution

1. Step 1: Recognize importance of data quality

   AI needs clean, consistent data to make accurate forecasts.

2. Step 2: Combine AI with human expertise

   Human insights help interpret AI results and improve decisions.

3. Final Answer:

   Clean and organize the data, then combine AI predictions with expert human insights -> Option D

4. Quick Check:

   Good data + human insight = better AI forecasts [OK]

Hint: Clean data and expert input improve AI forecasts [OK]

Common Mistakes:

• Using AI with bad data
• Ignoring human expertise
• Deleting useful historical data