Financial data analysis pattern in Data Analysis Python - Step-by-Step Execution

Concept Flow - Financial data analysis pattern

Load financial data

↓

Clean and prepare data

↓

Calculate key metrics

↓

Visualize trends and patterns

↓

Make data-driven decisions

This flow shows the main steps in analyzing financial data: loading, cleaning, calculating metrics, visualizing, and deciding.

Execution Sample

Data Analysis Python

import pandas as pd

# Load data
fin_data = pd.read_csv('financial_data.csv')

# Calculate daily returns
fin_data['Return'] = fin_data['Close'].pct_change()

# Calculate moving average
fin_data['MA_5'] = fin_data['Close'].rolling(window=5).mean()

This code loads financial data, calculates daily returns, and computes a 5-day moving average.

Execution Table

Step	Action	Variable/Column	Value/Result
1	Load CSV file	fin_data	DataFrame with columns: Date, Open, High, Low, Close, Volume
2	Calculate daily returns	fin_data['Return']	NaN for first row, then percent change of Close price
3	Calculate 5-day moving average	fin_data['MA_5']	NaN for first 4 rows, then average of last 5 Close prices
4	Check first 7 rows	fin_data.head(7)	Shows Date, Close, Return, MA_5 columns with calculated values
5	Exit	-	All calculations done

💡 All financial metrics calculated for the dataset

Variable Tracker

Variable	Start	After Step 2	After Step 3	Final
fin_data['Return']	Not present	NaN, 0.01, -0.005, 0.02, ...	Same as after Step 2	Final daily returns with NaN for first row
fin_data['MA_5']	Not present	Not present	NaN, NaN, NaN, NaN, 100.5, 101.2, ...	Final 5-day moving averages with NaN for first 4 rows

Key Moments - 2 Insights

Why is the first value of daily returns NaN?

Because there is no previous day to compare for the first row, so pct_change() returns NaN as shown in execution_table step 2.

Why do the first 4 values of the moving average column show NaN?

A 5-day moving average needs 5 data points, so the first 4 rows don't have enough data, resulting in NaN as seen in execution_table step 3.

Visual Quiz - 3 Questions

Look at the execution table, what is the value of fin_data['Return'] at the first row?

A NaN

B 0

C 1

D -1

💡 Hint

Refer to execution_table row 2 where daily returns are calculated and first value is NaN.

At which step do we first see the 5-day moving average values calculated?

A Step 1

B Step 2

C Step 3

D Step 4

💡 Hint

Check execution_table row 3 describing calculation of MA_5.

If the window size for moving average changed to 3, how would the NaN count in fin_data['MA_5'] change?

A More NaNs at start

B Fewer NaNs at start

C No NaNs at all

D NaNs remain the same

💡 Hint

Refer to key_moments about how window size affects initial NaNs in moving average.

Concept Snapshot

Financial data analysis pattern:
1. Load data (CSV or other)
2. Clean and prepare (handle missing data)
3. Calculate metrics (returns, moving averages)
4. Visualize trends (charts, graphs)
5. Use insights to decide

Full Transcript

This lesson shows how to analyze financial data step-by-step. First, we load the data from a CSV file into a DataFrame. Then, we calculate daily returns using percent change of closing prices. Next, we compute a 5-day moving average to smooth price trends. We observe that the first daily return is NaN because there is no previous day to compare. Similarly, the first 4 moving average values are NaN because we need 5 days of data to calculate it. These steps help us understand price movements and prepare for visualization or decision-making.

Practice

1. What is the main purpose of analyzing financial data in Python?

easy

A. To write novels

B. To create video games

C. To design websites

D. To understand money trends and make informed decisions

Solution

1. Step 1: Understand the goal of financial data analysis

   Financial data analysis focuses on studying money-related data to find patterns and trends.

2. Step 2: Match the goal with the options

   Only To understand money trends and make informed decisions talks about understanding money trends and making decisions, which fits the purpose.

3. Final Answer:

   To understand money trends and make informed decisions -> Option D

4. Quick Check:

   Financial data analysis = Understand money trends [OK]

Hint: Focus on money trends and decisions for financial analysis [OK]

Common Mistakes:

• Confusing financial analysis with unrelated fields
• Ignoring the purpose of data analysis
• Choosing options unrelated to money or data

2. Which of the following is the correct way to import pandas for financial data analysis in Python?

easy

A. import pandas as pd

B. import panda as pd

C. import pd as pandas

D. import pandas

Solution

1. Step 1: Recall the standard pandas import syntax

   The common way to import pandas is using 'import pandas as pd' for easy reference.

2. Step 2: Check each option for correctness

   import pandas as pd matches the standard syntax. Options B and C have wrong module names or order. import pandas imports pandas but without alias, which is less common.

3. Final Answer:

   import pandas as pd -> Option A

4. Quick Check:

   Standard pandas import = import pandas as pd [OK]

Hint: Use 'import pandas as pd' for standard data analysis [OK]

Common Mistakes:

• Misspelling 'pandas' as 'panda'
• Reversing import order
• Not using alias 'pd' which is common practice

3. Given the code:

import pandas as pd

data = {'Date': ['2024-01-01', '2024-01-02'], 'Price': [100, 105]}
df = pd.DataFrame(data)
df['Date'] = pd.to_datetime(df['Date'])
df.set_index('Date', inplace=True)
print(df.loc['2024-01-02'])

What will be the output?

medium

A. Error: KeyError

B. Price 105
Name: 2024-01-02 00:00:00, dtype: int64

C. Date 2024-01-02
Price 105

D. Price 100
Name: 2024-01-01 00:00:00, dtype: int64

Solution

1. Step 1: Understand DataFrame creation and index setting

   The DataFrame has dates converted to datetime and set as index. So rows are accessed by date index.

2. Step 2: Access row with loc for '2024-01-02'

   Using df.loc['2024-01-02'] returns the row with Price 105 and index as Timestamp '2024-01-02 00:00:00'.

3. Final Answer:

   Price 105 Name: 2024-01-02 00:00:00, dtype: int64 -> Option B

4. Quick Check:

   df.loc['2024-01-02'] = Price 105 [OK]

Hint: Remember loc uses index; datetime index returns full timestamp row [OK]

Common Mistakes:

• Expecting a DataFrame instead of a Series output
• Not converting 'Date' to datetime before indexing
• Using wrong date format causing KeyError

4. Identify the error in this code snippet for calculating daily returns:

import pandas as pd
prices = pd.Series([100, 105, 103, 108])
returns = prices.pct_change()
print(returns)

medium

A. No error; code correctly calculates daily returns

B. pct_change() should be pct_change(1)

C. prices should be a DataFrame, not Series

D. Missing import for numpy

Solution

1. Step 1: Understand pct_change() method

   pct_change() calculates percentage change between current and prior element by default (periods=1).

2. Step 2: Check code correctness

   prices is a Series, which supports pct_change(). No import or DataFrame needed. Code runs fine and prints daily returns.

3. Final Answer:

   No error; code correctly calculates daily returns -> Option A

4. Quick Check:

   pct_change() default periods=1 works [OK]

Hint: pct_change() defaults to 1 period; no extra args needed [OK]

Common Mistakes:

• Thinking pct_change needs argument 1 explicitly
• Confusing Series with DataFrame necessity
• Assuming numpy import is required for pct_change

5. You have a DataFrame df with columns Date (string), Open, and Close prices. How do you create a new column Daily_Return that shows the percentage change from Open to Close for each day?

hard

A. df['Daily_Return'] = df['Close'].pct_change()

B. df['Daily_Return'] = df['Open'] / df['Close']

C. df['Daily_Return'] = (df['Close'] - df['Open']) / df['Open'] * 100

D. df['Daily_Return'] = df['Close'] - df['Open']

Solution

1. Step 1: Understand daily return calculation

   Daily return is the percent change from Open to Close price: ((Close - Open) / Open) * 100.

2. Step 2: Match formula with options

   df['Daily_Return'] = (df['Close'] - df['Open']) / df['Open'] * 100 correctly applies the formula. df['Daily_Return'] = df['Close'].pct_change() calculates percent change between rows, not within a row. Options C and D do not calculate percentage change correctly.

3. Final Answer:

   df['Daily_Return'] = (df['Close'] - df['Open']) / df['Open'] * 100 -> Option C

4. Quick Check:

   Daily return = ((Close - Open)/Open)*100 [OK]

Hint: Use (Close - Open) / Open * 100 for daily percent return [OK]

Common Mistakes:

• Using pct_change() which compares different rows
• Dividing Open by Close instead of difference over Open
• Not multiplying by 100 for percentage