FIN-Econ-FF3-Sorting-Strategy

Overview

FIN-Econ FF3 Chinese Sorting Trading Strategies To Select Portfolios

This program is part of a course project for FIN ECON, demonstrating a trading strategy that leverages the Fama-French three-factor model (FF3), sorted by Earning/Market values (EM) and Free-Cash-Flow (FCF).

Highlights:

• Target: Chinese STSE stock market
• Annual Return: ~30% (historical performance)
• Accessibility: Easily replicable even for beginners
• Purpose: Exchange of trading ideas among students

⭐ If you find this useful (e.g., for homework, research, or trading), please support by starring this repository! 🌟🌟🌟🌟🌟🌟

---

Project Link

You can access the full project and its source code on GitHub:
👉 FIN-Econ-FF3-Sorting-Strategy 👈

---

Citation

If you use this project in your research, homework, or any other work, please make sure to cite it as follows:

BibTeX Citation Format:

@misc{FIN-Econ-FF3-Sorting-Strategy,
  author = {Fisher669},
  title = {FIN-Econ-FF3-Sorting-Strategy: Trading Strategies Sorted by Earning/Market Values and Free-Cash-Flow},
  year = {2024},
  howpublished = {\url{https://github.com/Fisher669/FIN-Econ-FF3-Sorting-Strategy/}},
  note = {Accessed: [Insert Date]}
}

APA Citation Format:

Fisher669 (2024). FIN-Econ-FF3-Sorting-Strategy: Trading Strategies Sorted by Earning/Market Values and Free-Cash-Flow. Retrieved from https://github.com/Fisher669/FIN-Econ-FF3-Sorting-Strategy/

By citing this work, you help support open-source development and academic exchange of ideas!

---

Introduction

Motivation

1. Why FF3 Trading Strategies?

   • FF3 is a widely recognized model in financial economics that incorporates three key factors: market, size, and value. It provides a robust framework for understanding excess returns.

2. Why Earning/Market Values (EM) and FCF for Sorting?

   • EM = 1/PE ratio (Earning/Market Share) represents the forward-looking market sentiment, while FCF (Free-Cash-Flow) reflects a company's internal financial health.

3. Why Combine EM and FCF?

   • Combining these two criteria helps balance market-driven expectations with fundamental financial strength.
   • Investigating their correlation and synergy is crucial for robust portfolio selection.

---

Model

Data Selection

• Source: CSMAR (China Securities Market Analysis and Research)
• Date Range: 2014-01-01 to 2023-12-31
• Frequency: Monthly
• Coverage: Equities, Fixed Income, Commodities, and FX
• Missing Data Handling: Quarterly Rebalance so we choose to use end of Q. to impute this.

---

Main Sorting Criteria

1. Earning/Market values (EM):

• Represents growing company probabilities.
• Formula:
  EM = Earning/Market Value

2. Free-Cash-Flow (FCF):

• Measures the cash flow available to investors after accounting for obligations.
• Formula:
  Total Assets - Total Liabilities

---

Strategy Selection

Why FF3?

The FF3 model adds granularity to portfolio selection through the following factors:

1. Market Factor: Captures the overall market return.
2. Size Factor: Differentiates between small-cap and large-cap stocks.
3. Value Factor: Identifies undervalued companies with high book-to-market ratios.

---

Results

This section summarizes the outcomes of the strategy selection and evaluation.

Metrics:

1. Average Monthly Raw Return:

   • Presented in Table 1.

2. Average Monthly Return (with t-stat):

   • Presented in Table 2.

Ablation Test

• Individual Sorting:

  • Results for strategies sorted by FCF only and EM only.
  • Two tables (similar format to Table 1 & 2).

• Combined Sorting:

  • Results when both FCF and EM are combined.
  • Comparison with individual sorting performance.

Results Report

Main Results

Exhibit 1: Monthly Raw Return for 25 Stock Portfolios Formed on FCF and E/M (2014-2023)

Earnings/Market Value (E/M) Quintile	1 (Low)	2	3	4	5 (High)
FCF Quintile 1 (Low)	0.0254	0.0180	0.0046	-0.0001	0.0007
FCF Quintile 2	0.0141	0.0185	0.0110	0.0034	0.0003
FCF Quintile 3	0.0138	0.0170	0.0087	0.0060	0.0021
FCF Quintile 4	0.0226	0.0198	0.0091	0.0048	0.0012
FCF Quintile 5 (High)	0.0238	0.0147	0.0085	0.0066	0.0033

Exhibit 2: Intercepts from Excess Stock Return Regression for 25 Stock Portfolios Formed on FCF and E/M

Earnings/Market Value (E/M) Quintile	Alpha	t(a)
FCF Quintile 1 (Low)	0.0155	5.39
	0.0084	2.72
	-0.0047	-1.38
	-0.0094	-4.77
	-0.0083	-4.60
FCF Quintile 2	0.0035	1.82
	0.0081	3.04
	0.0016	0.67
	-0.0058	-3.05
	-0.0084	-5.69
FCF Quintile 3	0.0029	1.56
	0.0074	3.44
	-0.0008	-0.45
	-0.0032	-2.11
	-0.0074	-4.16
FCF Quintile 4	0.0126	4.64
	0.0098	3.64
	-0.0005	-0.24
	-0.0045	-2.35
	-0.0077	-4.94
FCF Quintile 5 (High)	0.0142	6.33
	0.0061	3.30
	-0.0005	-0.25
	-0.0024	-1.35
	-0.0056	-2.72

Additional Tests

(1) Independent Sorting

Exhibit 3: Monthly Raw Return for 25 Stock Portfolios Formed on FCF and E/M (Independent Sorting)

Earnings/Market Value (E/M) Quintile	1 (Low)	2	3	4	5 (High)
FCF Quintile 1 (Low)	0.0256	0.0199	0.0080	-0.0012	0.0010
FCF Quintile 2	0.0143	0.0168	0.0069	0.0034	-0.0021
FCF Quintile 3	0.0157	0.0127	0.0060	0.0041	0.0026
FCF Quintile 4	0.0227	0.0193	0.0106	0.0042	0.0013
FCF Quintile 5 (High)	0.0209	0.0223	0.0137	0.0071	0.0051

Exhibit 4: Intercepts from Excess Stock Return Regression for 25 Stock Portfolios Formed on FCF and E/M (Independent Sorting)

Earnings/Market Value (E/M) Quintile	Alpha	t(a)
FCF Quintile 1 (Low)	0.0155	4.77
	0.0104	3.24
	-0.0013	-0.38
	-0.0103	-5.46
	-0.0081	-4.88
FCF Quintile 2	0.0038	1.99
	0.0068	2.63
	-0.0026	-1.27
	-0.0056	-3.57
	-0.0105	-6.74
FCF Quintile 3	0.0051	3.05
	0.0033	1.55
	-0.0032	-1.78
	-0.0054	-3.40
	-0.0067	-3.07
FCF Quintile 4	0.0128	4.56
	0.0094	4.32
	0.0007	0.33
	-0.0052	-2.95
	-0.0077	-4.90
FCF Quintile 5 (High)	0.0109	4.05
	0.0136	5.11
	0.0050	2.83
	-0.0019	-1.12
	-0.0038	-2.23

Reverse Test

Exhibit 5: Monthly Raw Return for 25 Stock Portfolios Formed on FCF and E/M (Reverse Sorting)

FCF Quintile	1 (Low)	2	3	4	5 (High)
E/M Quintile 1 (Low)	0.0233	0.0141	0.0141	0.0191	0.0237
E/M Quintile 2	0.0206	0.0174	0.0121	0.0167	0.0228
E/M Quintile 3	0.0094	0.0059	0.0062	0.0106	0.0137
E/M Quintile 4	-0.0009	0.0037	0.0031	0.0042	0.0075
E/M Quintile 5 (High)	0.0005	0.0006	0.0006	0.0030	0.0065

Exhibit 6: Intercepts from Excess Stock Return Regression for 25 Stock Portfolios Formed on FCF and E/M (Reverse Sorting)

Free Cash Flow (FCF) Quintile	Alpha	**t
E/M Quintile 1 (Low)	0.0098	3.34
	0.0030	1.03
	-0.0014	-0.50
	-0.0044	-1.67
	-0.0061	-2.88
E/M Quintile 2	0.0043	2.56
	0.0033	1.94
	-0.0029	-1.49
	-0.0044	-2.26
	-0.0075	-3.91
E/M Quintile 3	0.0016	1.03
	0.0008	0.44
	-0.0036	-2.06
	-0.0049	-3.13
	-0.0062	-3.41
E/M Quintile 4	0.0060	2.85
	0.0024	1.15
	-0.0020	-1.07
	-0.0039	-2.17
	-0.0055	-3.29
E/M Quintile 5 (High)	0.0050	2.74
	0.0070	3.78
	0.0040	2.12
	-0.0010	-0.57
	-0.0024	-1.36

---

You can now copy and paste this into your GitHub repository. If you need further refinements or adjustments, feel free to let me know!

---

Conclusion

This project demonstrates the utility of sorting trading strategies using FF3 factors combined with EM and FCF. The results show that this approach can achieve notable returns in the Chinese STSE stock market.

Whether you're a student, researcher, or investor, this repository serves as a foundation for exploring FF3-based trading strategies.

---

Acknowledgements:
This project is for academic purposes only and is intended to foster idea exchange among students.

Star this repo if you found it helpful! 🌟