{ "cells": [ { "cell_type": "markdown", "id": "08b6abf2-245e-4217-86fb-536e3d6b0241", "metadata": {}, "source": [ "---\n", "\n", "**Assignment: Financial Research Control Variables Using Python**\n", "\n", "**Objective:**\n", "In this assignment, you will compute a set of commonly used control variables in financial research using Python. The original data comes from the CSMAR (China Stock Market & Accounting Research) database. You are expected to implement the necessary data processing, calculations, and summarize your findings effectively.\n", "\n", "---\n", "\n", "### **Instructions:**\n", "\n", "1. **Download and Load Data**\n", "\n", " * Obtain the raw data from the CSMAR database. The data should contain financial information for multiple companies across various years(2003-2024). For the sake of this assignment, use data that includes key variables such as **Revenue**, **Total Assets**, **Equity**, **Operating Profit**, **Cash Flow**, and others relevant for financial analysis.\n", "\n", "2. **Control Variables Calculation**\n", " For each company-year observation in the dataset, calculate the following control variables, which are commonly used in financial research:\n", "\n", " * **Size**: The natural logarithm of Total Assets.\n", "\n", " ```python\n", " df['Size'] = np.log(df['Total_Assets'])\n", " ```\n", " * **Leverage (Lev)**: Total Debt / Total Assets.\n", "\n", " ```python\n", " df['Lev'] = df['Total_Debt'] / df['Total_Assets']\n", " ```\n", " * **Profitability (ROA)**: Return on Assets = Net Income / Total Assets.\n", "\n", " ```python\n", " df['ROA'] = df['Net_Income'] / df['Total_Assets']\n", " ```\n", " * **Growth**: The annual percentage change in Revenue.\n", "\n", " ```python\n", " df['Growth'] = df['Revenue'].pct_change() * 100\n", " ```\n", " * **Market-to-Book Ratio (MBR)**: Market Value of Equity / Book Value of Equity.\n", "\n", " ```python\n", " df['MBR'] = df['Market_Capitalization'] / df['Total_Equity']\n", " ```\n", " * **Cash Flow**: Operating Cash Flow / Total Assets.\n", "\n", " ```python\n", " df['CashFlow'] = df['Operating_Cash_Flow'] / df['Total_Assets']\n", " ```\n", "\n", "3. **Additional Control Variables**\n", " Add the following control variables that are also commonly used in financial research:\n", "\n", " * **State-Owned Enterprise (SOE)**: A binary variable indicating whether the company is state-owned. You can assume that a variable `SOE` is already provided in the dataset, with `1` for state-owned enterprises and `0` for others.\n", "\n", " ```python\n", " df['SOE'] = df['SOE'] # Use existing data, or create if necessary.\n", " ```\n", " * **Dual Role (DUAL)**: A binary variable indicating whether the CEO also serves as the Chairman of the Board. You may need to create this variable based on the dataset.\n", "\n", " ```python\n", " df['DUAL'] = df['CEO_Chairman'] # Create if applicable.\n", " ```\n", " * **First Shareholder Ownership Ratio (H1)**: The percentage of shares held by the largest shareholder.\n", "\n", " ```python\n", " df['H1'] = df['H1_Shareholding']\n", " ```\n", " * **Management Shareholding Ratio (MSH)**: The percentage of shares held by the management team.\n", "\n", " ```python\n", " df['MSH'] = df['Management_Shareholding']\n", " ```\n", " * **Institutional Investor Shareholding Ratio (INSTSH)**: The percentage of shares held by institutional investors.\n", "\n", " ```python\n", " df['INSTSH'] = df['Institutional_Shareholding']\n", " ```\n", " * **Independent Directors Ratio (INDP)**: The ratio of independent directors on the board.\n", "\n", " ```python\n", " df['INDP'] = df['Independent_Directors'] / df['Board_Size']\n", " ```\n", " * **Board Size (BSIZE)**: The total number of board members.\n", "\n", " ```python\n", " df['BSIZE'] = df['Board_Size']\n", " ```\n", "\n", "4. **Data Processing**\n", "\n", " * Ensure that the dataset has no missing values in the required columns. Use appropriate methods to handle missing values (e.g., imputation, deletion, etc.).\n", " * Convert variables to appropriate data types if necessary (e.g., converting year columns to integers).\n", "\n", "5. **Create New Columns**\n", " Create new columns for each calculated control variable. Make sure to use meaningful names and ensure no overwriting of the original columns unless explicitly required.\n", "\n", "6. **Data Summary**\n", " After processing the data and calculating the control variables, provide a summary of the dataset:\n", "\n", " * Number of companies and years represented.\n", " * A brief descriptive analysis of the control variables (mean, median, standard deviation).\n", "\n", " Example:\n", "\n", " ```python\n", " df[['Size', 'Lev', 'ROA', 'Growth', 'MBR', 'CashFlow', 'SOE', 'DUAL', 'H1', 'MSH', 'INSTSH', 'INDP', 'BSIZE']].describe()\n", " ```\n", "\n", "7. **Save Processed Data**\n", " Save your processed dataset to a new file (e.g., `processed_data.csv` or `processed_data.dta`) and ensure the control variables are correctly included.\n", "\n", "---\n", "\n", "### **Deliverables:**\n", "\n", "1. Python code implementing all the steps above.\n", "2. A summary report that includes:\n", "\n", " * A description of the dataset and the steps taken to clean and process the data.\n", " * A table showing the descriptive statistics for the control variables.\n", " * Any additional observations or issues encountered during the analysis.\n", "\n", "---\n", "\n", "**Submission Deadline:** [2026-04-12]\n", "\n", "**Submission Format:** Please submit a Python script (`.py`) and a report in a PDF format.\n", "\n", "Good luck, and make sure to follow the instructions carefully!" ] }, { "cell_type": "code", "execution_count": null, "id": "8d5ab9a9-1454-4f89-8784-f0990452e6fe", "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.14.3" } }, "nbformat": 4, "nbformat_minor": 5 }