MINISTRY OF EDUCATION AND TRAINING STATE BANK OF VIETNAM BANKING UNIVERSITY OF HO CHI MINH CITY BACHELOR THESIS Major Financial – Banking Number 7340201 Topic APPLICATION OF FAMA FINANCIAL MODEL TO IN[.]
INTRODUCTION
Reason to research
The financial exchange and banking sector play a vital role in the national economy, with investors—both institutional and individual—seeking optimal returns on their investments Selecting stocks can resemble gambling, where understanding market dynamics is crucial for identifying profitable opportunities Share prices fluctuate to reflect market valuation, impacting profit margins and necessitating a thorough analysis of pricing trends, risks, historical performance, and unpredictable future outcomes Investors prioritize assessing the potential success of their investments before purchasing, making it essential to grasp various fundamental forces that influence market behavior Over the past century, extensive research has led to the development of numerous pricing models, including the Capital Asset Pricing Model (CAPM), which has been foundational in understanding investment dynamics since its introduction in the 1960s.
The Capital Asset Pricing Model (CAPM), established in 1966, relies solely on beta as the market risk factor to predict stock returns, yet its reliability has faced significant criticism Basu (1977) highlighted the model's inadequacies in the Indian context, while Rolf W Banz (1981) identified its misspecification, leading to broader acceptance of its limitations for NYSE stocks This prompted Fama and French to explore the connections between stock returns and various factors, resulting in the introduction of the Fama-French three-factor model, which gained popularity in the 1980s and eventually replaced the CAPM The model's effectiveness was validated across global markets, including Australia, Canada, and Europe In 1997, Mark Carhart enhanced this framework by introducing a four-factor model that incorporated a momentum factor for asset valuation, which also serves as a tool for mutual fund evaluation Subsequent research by Novy-Marx (2013) and Aharoni et al (2013) underscored the importance of earnings and spending on profitability Building on these findings, Fama and French proposed a five-factor model in 2015 aimed at optimizing financial decision-making by focusing on conservative versus aggressive investments, demonstrating success in various regions, including North America and Japan.
The Fama five-factor model is gaining significant attention from investors, particularly in the equity market, yet many researchers have yet to address the underlying issues explicitly Vo Hong Duc and Mai Duy Tan (2014) conducted an analysis that involved grading portfolios through various regression models and categorizing them based on their findings However, replicating these portfolios can yield unexpected results due to the interconnections among different variable sets Moreover, modeling portfolios based solely on 14 individuals may not be sufficient for establishing credibility.
The article explores the application of Fama-French factors within the industrial sector of the Vietnam stock market It aims to introduce this investment concept to local investors, providing insights that can help maximize their value and enhance their stock market strategies.
Research objective
The aim of this thesis was to:
Firstly, analyze the influence of the five-factor model, including industry, scale, valuation, benefit, and investment factors has on listed industrial stocks returns in the Vietnam stock market
Secondly, describe the relevant valuation model and the fluctuation of the Vietnamese capital market returns in a simple and detailed manner
Finally, offer several ideas on how owners, regulators, and other stockholders may enhance the continuing management of the fund.
Research questions
To accomplish the above study's purpose, these are the questions it seeks to address:
The book-to-market ratio, profitability, scale, market premium, and investment risk significantly influence a portfolio's returns These external factors can create either a favorable or negative relationship with stock performance, impacting overall investment outcomes Understanding these connections is crucial for optimizing portfolio strategies and enhancing returns.
- The Fama French five-factor model is sufficient method for describing the shifts in returns in the equity market in Viet Nam?
- Why investors make use of analysis to raise equity capital and reduce investment risks?
Research subject and range
The study emphasis is on utilizing the Fama French Five-Factor Pricing Model for mentioned manufacturing firms on the HNX and HOSE exchanges
- The time frame for the study is from 2014 to 2019 Prioritizing the objective to create an accurate analysis, any earlier return data is disregarded in this study
The firms included in the study are primarily focused on the Industrials sector, must be publicly listed, and are required to provide readily available data on key financial metrics such as Market Price, Total Assets, Total Liabilities, Shares Outstanding, Book Value, and Treasury bills, as sourced from the VNCB over a three-month survey period.
This analysis focuses on the reported market capitalization of industrial firms listed on HOSE and HNX, excluding companies from the banking sector, such as insurance firms and brokerage companies.
Methodlogy
The aim of the analysis was to evaluate the Fama French Five Factor Model in Vietnamese industrial firms, a quantitative methodology was implemented:
- Follow the Ordinary Least Square (OLS) procedure to quantify the Betas, and analyze the association between variables and portfolios
- Using Gibbons, Ross, and Shanken (1989) GRS model to approximate the fundamental influence of the model on the list of firms
- Excel Office is used to synthesize data and equations accompanied by the usage of Stata version 13 to execute regression and other related hypothesis testing procedures
The expected return on asset i is influenced by the risk-free rate of Treasury bills, the excess market return, and various factors including the size factor (Small minus Big), the value factor (High minus Low), the profitability factor (Robust minus Weak), and the investment factor (Conservative minus Aggressive) Additionally, the asset's sensitivity is represented by the coefficients, while the intercept and error term at time t further define the asset's return dynamics.
Research contribution
The thesis provides many unique contributions:
This study aims to validate the usability of the Fama French five-factor pricing models, providing clarity on its factors for investors and researchers focused on predicting future income rates while minimizing immediate risks Consequently, this concept can be directly applied to the Vietnamese capital market.
Experimentally, by assessing the feasibility of the templates, analyzing the test findings, and presenting any hints to investors and individuals when choosing and handling the portfolio.
Research outline
This chapter introduces the motives for conducting this project, the research aims, the research subject, the research range and the scope of work
This chapter presents the theoretical background behind the current study, and previous research into a similar subject
This chapter details the study's framework and experimental specifics, outlining the dependent and independent variables It provides guidance on portfolio construction and explains regression analysis along with the necessary steps for its application.
This chapter presents a regression analysis that illustrates the impact of the primary model discussed in Chapter 3 It encompasses data on all relevant variables, featuring associations, graphical representations, and a comparative analysis of different models Each segment concludes with a summary of the findings and references to prior research.
This study proves valuable by providing a comprehensive interpretation of the analysis, along with actionable insights for business owners, banking professionals, and policymakers It also addresses the limitations of the current analysis and suggests directions for future research.
LITERATURE REVIEW AND PREVIOUS RESEARCHES
Literature review
In 1976, Ross introduced the Arbitrage Pricing Theory (APT), a groundbreaking concept that significantly influenced asset valuation This theory posits that the expected return on a financial asset can be modeled as a linear function of various macroeconomic variables or theoretical market indices APT has gained immense popularity and is utilized in the trading of stocks and goods across different markets, facilitating arbitrage opportunities through currency exchanges.
However, it is not a model, but rather a simplified hypothesis of financial returns The Expected Return of a stock i is a function that represents both systemic and non- systematic risk factors
Where: is the expected return on asset i
The risk-free interest rate in government bonds serves as a benchmark for assessing the asset beta, which reflects the sensitivity of various risk factors Additionally, the risk premium associated with each factor, denoted as k (where k = 1, 2, n), plays a crucial role in determining the overall risk profile of the stock, represented by the variable i.
Diversification of a portfolio can significantly mitigate non-systematic threats, while compensatory considerations are primarily linked to systematic risks Systematic risk factors are encompassed within the framework of the Arbitrage Pricing Theory (APT) hypothesis.
- Evaluates a difference between short-term and long-term interest rate
- How different government and business bonds vary
When comparing Arbitrage Pricing Theory (APT) and Capital Asset Pricing Model (CAPM), APT emerges as a more flexible approach, as it does not necessitate the existence of a stock portfolio and allows for the identification of mispriced specific stocks Unlike CAPM, which is often criticized for its unrealistic assumptions, APT accommodates a diverse range of variables in its multifactor models, reflecting the uncertainty surrounding the number and nature of risk factors Although CAPM is frequently favored for its versatility and broad applicability across sectors, research has proposed alternative asset valuation methods that challenge its foundational assumptions Ultimately, while APT offers a framework that incorporates macroeconomic elements to assess expected returns, it presents challenges in selecting the appropriate variables for its model.
2.1.2 The Fama French three-factor model
In the 1990s, the capital asset pricing model (CAPM) was a significant focus for scholars, recognized for its broad business applications However, it ultimately failed as a historical asset pricing model, leading Eugene Fama and Kenneth French to advocate for a non-beta model that better explained the data Building on William Sharpe's work, their analysis integrated multiple factors influencing predicted returns, such as company size, financial leverage, E/P ratio, BE/ME ratio, and stock beta They found that the relationship between beta and standard deviation did not adequately account for average stock returns from the 1960s to the 1980s This prompted a search for additional factors that could more effectively justify stock returns beyond the single variable of sector beta used in earlier models Subsequent research highlighted the importance of firm size and book-to-market ratio, revealing their strong correlation with equity returns, while the impact of P/E and financial leverage was less clear when included in the model.
Fama and French (1993) conducted an analysis comparing stocks with limited capital market value and those with broader value, revealing that the size and value factors significantly influenced market price movements more than the beta factor when included in their regression model Their findings utilized two variables, scale and meaning, to illustrate the impact of these factors on portfolios, suggesting that their regression model effectively describes equity prices.
Where: is the expected return of asset i at time t is the risk-free interest rate of government bonds is the excess market return
(Small minus Big) the size risk factor
(High minus Low) the value risk factor
The coefficients , 𝑠 , and are the asset’s sensibility is the constants intercept is the error term at time 𝑡
The Fama French model demonstrates that investors who embrace higher risks can achieve greater returns This analysis highlights the impact of the SMB and HML variables on the profitability of portfolio i, which consists of stocks characterized by strong growth potential and low risk Portfolio i encompasses valuable stocks with high growth and low volatility, while also considering factors related to the financial sector and equity market dynamics.
This model effectively summarizes findings from previous studies, including well-known CAPM analyses, by utilizing observational data collected from various regions such as South Africa, India, Ukraine, and Taiwan Consistent with the Fama French three-factor model, portfolio returns have shown reliability; however, some researchers argue that these three variables alone cannot fully determine systematic risk premiums and suggest the presence of additional factors influencing profitability Notably, Novy-Marx (2013) demonstrated that enhanced gross profitability better explains stock return differences than the book-to-market ratio, while Hou et al (2015) found that investment and return levels account for variance in stock performance.
Nartea et al (2009) found that the Carhart four-factor model effectively explained momentum returns, unlike the Fama-French model, which focuses on three factors Carhart's model introduces a momentum factor, defined as the return difference between high-performing stocks (winners) and low-performing stocks (losers) over the past 3 to 12 months Jegadeesh and Titman (2001) support this by suggesting that buying winning stocks and selling losing ones can enhance overall returns This phenomenon occurs as investors often sell to realize gains, anticipating better opportunities Additionally, Carhart (1997) enhances the Fama-French model by incorporating an anomaly factor that compares returns of one-year winners and losers, termed 'Robust minus Weakness' (RMW) The WML factor, similar to the HML factor, analyzes stock performance from the previous year, highlighting that small-cap stocks do not exhibit the same patterns as large-cap stocks, leading to a complex outcome.
𝑟 is the expected return on asset i
𝑟 is the risk-free interest rate of government bond
(Small minus Big) the size risk factor
(High minus Low) the value risk factor
(Winner minus Loser) the momentum risk factor
The coefficients , is the asset’s sensibility is the constants intercept the error term of asset i at time t
The four-factor approach has been successfully applied to various established economies, including the United States and Europe, demonstrating a better fit for data compared to the three-factor model Wei Zhang (2018) enhanced the Carhart four-factor model, revealing that reversal effects are not accounted for by the Fama-French three-factor model, while providing superior explanations for the relationship between Chinese stock returns and historical data, along with alternative investment strategies Additionally, the French and Fama (2014) model incorporates five factors, further expanding the framework for understanding stock returns.
2.1.4 The Fama French five factor model
Where: is the share price at time t
E(dt+τ) is the expected dividend per share in period t+τ r is (approximately) the long-term average expected stock return (the internal rate of return on expected dividends)
Miller & Modigliani (1961) suggest the overall market valuation of the firm's portfolio at time t by (4) as following:
, is total equity earnings for period 𝑡 the change in total book equity Dividing by the time t book equity gives:
The dividend discount model indicates that stocks with strong profitability typically offer higher projected returns compared to those with weak profitability Fama-French (2015) enhanced their original three-factor model by incorporating two additional variables: the profitability factor (RMW – Robust minus Weak) and the investment factor (CMA – Cautious minus Aggressive) This development led to the creation of a 5-factor model, which is poised to serve as the new standard for asset pricing research.
𝑟 is the expected return on porfolio i for period t
𝑟 is the risk-free interest rate of government bonds for period t
𝑟 𝑟 is the excess market return for period t
(Small minus Big) the size risk factor
(High minus Low) the value risk factor
(Robust minus Weak) the profitability risk factor
(Conservative minus Aggressive) the investment risk factor
The coefficients 𝑠 𝑟 are the portfolio’s sensibility is the constants intercept is the error term of asset i at time t
A study by Stephen Cochrane reveals that expenditure and profitability factors are more affected by economic conditions than scale factors, highlighting their significance in assessing hedge fund strategies' performance throughout market cycles Many hedge funds aim to exploit risk premiums associated with market anomalies, such as the tendency for small businesses to outperform the market This asset pricing approach primarily relies on data from developed countries, particularly the United States Additionally, research by Chan and Hamao (1991) emphasizes the importance of value in understanding Japanese stock market returns, contradicting Fama and French's (2010) conclusions Despite limited scientific research on the five-factor model in the Vietnamese stock market, there is evidence supporting its superiority over the three-factor model The upcoming chapter will delve deeper into the methodology and findings from four experiments related to this topic.
Previous researches
2.2.2 Previous researches from developed countries
Research in US of Ferson & Harvey (1999)
This research examines unconditional mean returns, highlighting that previous studies have attempted to characterize average returns It finds that autocorrelations of fund returns are typically low, around 0.1 for small portfolios, with some statistically significant exceptions The HML section does not allow for measuring estimated returns across different time horizons, and as a result, the coefficients on HML in regressions diminish, rendering t-ratios insignificant The business beta coefficient tends to be higher when there is a fit, while intercepts narrow significantly when the model is applied Overall, regression intercepts approach zero in the three-factor construct, and while alphas are present, they often vary over time The findings suggest that the Fama-French three-factor model fails to adequately describe the returns or Sharpe ratio of this portfolio, and a variant incorporating time-varying betas is also rejected.
Research in Japan of Daniel, Titman and Wei (2001)
The analytical study identified a significant association between average excess returns and ex-ante factor loading rankings, revealing that the Fama-French three-factor model tends to overestimate the relationship between stock market returns and various factors This discrepancy may stem from a low variance in the HML beta Additionally, the research demonstrated a consistent ordering of ex-ante HML factors and ex-post factor loadings, with the HML factor contributing 0.586 to individual stock portfolios, supported by a t-statistic of 14.14 The model predicts a zero intercept; however, the expected slope is negative 20.205, with standard errors of 1.80 from zero The minimal difference between the two portfolio returns suggests that the Fama-French model should be reconsidered.
Research in French of Souad Ajili (2002)
In a study examining the French economy from July 1991 to June 2001, Souad Ajili compared the Fama French three-factor model with the Capital Asset Pricing Model (CAPM) across various equity portfolios The analysis included equal-weight returns of commodities, value-weight returns of securities, and four indices: CAC40, SBF80, SBF120, and SBF250 The results suggest that the Fama French three-factor model is more effective in explaining common stock returns than CAPM, with a typical regression outcome of 0.905.
Research in US of Zhu (2016)
Zhu enhanced the Fama French five-factor model by incorporating the Standardized Standard Asymmetric Exponential Power Distribution (SSAEPD) and GARCH-type volatility to address existing gaps This study aims to determine if this expanded model outperforms the original Fama French five-factor framework Utilizing US stock data from July 1963 to December 2013, empirical evidence reveals that the five factors significantly influence asset valuation and cost assessment.
Research in Australia of Chiah and partners (2016)
This research analyzed a Fama-French portfolio of stocks over a 31-year period, revealing that scale, demand, rate of return, and profitability are interdependent factors Notably, in developing economies, the five-factor Fama-French model provides a more accurate depiction of returns compared to both the three-factor Fama-French model and the Capital Asset Pricing Model (CAPM).
Research in Japan of Keiichi Kubota and partner (2017)
In their long-run analysis of Japan's market, Keiichi Kubota and Hitoshi Takehara assert that market prices are well calibrated and challenge the effectiveness of the Fama and French five-factor model in differentiating Japanese data compared to traditional models They found that the RMW (Robust–minus–Weak) and CMA (Conservative–minus–Aggressive) factors did not serve as effective explanatory variables in generalized GMM tests using the Hansen–Jagannathan distance scale Their conclusion indicates that the three-factor model demonstrates comparable strength to the five-factor model, providing a clearer perspective on the data Notably, the four-factor model shows a robust test figure of 9.989, closely trailing the Capital Asset Pricing Model (CAPM) at 10.064, while the three-factor model outperforms all others in terms of performance.
2.2.3 Previous researches in developing countries
Research in European countries of Steven L, K Geert and Roberto (1999)
This research investigates the effectiveness of beta and the t-factor using CAPM and the Fama-French three-factor model to explain return variability across 12 European countries The authors evaluated accessible securities based on their betas, revealing that lower asset allocation investments yield the highest potential returns The findings indicate that the selected beta portfolios are indeed growing return portfolios, with t-values of 2.08 and 2.58 Additionally, each portfolio size exhibits a declining logarithmic trend, with the small firm portfolio showing a significant positive intercept of 0.62% per month (t=3.43) Although the intercepts appear minimal, the joint F-statistic from Gibbons et al (1989) strongly contradicts the assumption that these intercepts are negligible The beta and size-sorted portfolios demonstrate greater diversification than traditional national portfolios, suggesting that the return premium linked to size-based portfolios may stem from the inherent risk similarities among portfolios of varying sizes.
Research in China of Grace Xing Hu and partners (2018)
This analysis indicates a strong relationship between demand and firm size in China from 1990 to 2016, revealing that smaller businesses tend to outperform larger ones, leading to a reduction in uncertainties and an average return increase of 1.23% Utilizing the Fama-French methodology, the SMB factor demonstrates a significant monthly return of 0.61, indicating both statistical and economic relevance In contrast, the average returns of stocks show no consistent correlation with their book-to-market (B/M) ratios, as evidenced by the HML factor's average monthly return of 0.23, which, while positive, lacks significance The findings suggest no relationship between the demand parameter and premium, with SMB consistently exhibiting a positive coefficient in Fama-French cross-sectional analyses, making it crucial for understanding fluctuations in Chinese stock returns Furthermore, both studies highlight that early years of elevated uncertainty significantly influenced return variations, although this effect diminishes when analyzing long-term data corrections.
Research in India of Harshita and partners (2015)
An analysis of fifteen years of data on the CNX500 reveals a positive correlation in the Indian equity market between market capitalization and returns, profitability and returns, as well as the book-to-market (B/M) ratio and returns, supporting the findings of Fama and French.
The three-factor asset pricing model (CAPM), established in 1993, is optimal for single portfolios, while the five-factor model by Fama and French, introduced in 2015, is more effective for analyzing multiple portfolios This model yields the best results when the portfolio is devoid of any specific elements, making the five-factor approach more advantageous for asset investing.
Research in Turkey of Songul Kakilli Acaravci and partner (2017)
This research checked the validity of the five factor model by implementing it in Borsa Istanbul (BIST) during the 132-month period between July 2005 and June 2016 These
Fourteen intersection portfolios based on size, book-to-market (B/M) ratio, profitability, and valuation parameters were analyzed The GRS-F test confirmed the null hypothesis, supporting the validity of the consumption model Additionally, the five-factor approach appears applicable to the BIST, significantly impacting fund efficiency The mean average value observed in this model is 0.33, reinforcing the effectiveness of the Fama-French five-factor model in explaining excess portfolio returns.
Research of Truong Dong Loc and Duong Thi Hoang Trang (2014)
This research analyzes the extension of the Fama-French three-factor model to the HOSE stock market, utilizing data from January 2006 to December 2012 The findings reveal a positive correlation between the profitability of listed firms on HOSE and factors such as sector risk, company size, and book-to-market (B/M) ratio Business conditions significantly impact the profitability across all six portfolios examined Notably, the size factor shows a positive relationship with the profitability of small companies, while it negatively affects the returns of larger firms Additionally, the high minus low (HML) factor is positively associated with high and medium B/M ratio portfolios but negatively correlated with low B/M ratio portfolios Overall, the study confirms that the Fama-French three-factor model effectively explains the variations in profitability observed in HOSE indices.
Research of Vo Hong Duc và Mai Duy Tan (2014)
This report evaluates the Fama-French three-factor and five-factor models using data from 281 companies listed on the Ho Chi Minh City Stock Exchange between January 2007 and December 2015 The three-factor model highlights the demand factor as having a consistent positive impact, while also noting its original negative aspect, which remains statistically significant The size factor shows an optimistic trend, and the importance factor is statistically significant as well Additionally, profitability plays a crucial role, with a positive correlation to expenditures In conclusion, the findings underscore the relevance of these factors in investment analysis.
Fama French five-factor isn't sufficient to clarify return outcomes for Vietnam stock market
Research of Nguyen Thi Thuy Nhi (2016)
This research examines the Fama-French five-factor model and the Q-factor model by Hou, utilizing data from the HOSE and HNX stock markets between January 2009 and June 2015 The study employs three portfolio division strategies and finds that the demand effect is positive, the SMB factor is positive with limited portfolio scale, while the HML factor is negative for larger portfolios Additionally, the RMW factor shows a positive correlation with high ROE, and the CMA factor is positive with low operating profitability The regression model's explanatory power improved significantly, increasing from 80% to 96%, demonstrating that the Fama-French five-factor model provides a more comprehensive understanding than the Q-factor model.
Research of Huynh Ngoc Minh Tram (2017)
The analysis reveals that the SMB factor significantly enhances the importance of the HML factor and the MRP market return factor in predicting stock returns, with coefficients that are statistically significant at the 5% level Notably, only the SMB and MRP factors show positive expectations, while the HML factor remains nearly negative, indicating that smaller companies or those with lower book-to-market ratios can still generate income Additionally, the RMW and CMA factors are deemed unimportant, suggesting that the Fama French five-factor model does not fully account for investment returns in the Vietnamese stock market There exists a strong correlation between equity price variability, the market risk index, and stock returns in Vietnam.
DATA AND METHODOLOGY
Data construction and processing method
This research examines industries listed on the HOSE and HNX from 2014 to 2019, utilizing quarterly data from January 2014 to December 2019 The study focuses exclusively on newly listed manufacturing companies, excluding many firms due to poor financial performance and regulatory issues Ultimately, the analysis highlights the top 100 firms selected across this six-year period.
In 2017, Trinh Minh Quang utilized the Thomson Reuters database to analyze stock returns in Vietnam's industrial sector The study incorporated data from various Vietnamese newspapers, including Vietstock, Cafef, VnDirect, and Sbv, which were also featured in the annual reports of local projects The primary objective of this analysis was to assess the Fama-French five-factor model's effectiveness in evaluating stock returns for publicly listed industrial companies in the Vietnamese stock market.
The information regarding outstanding bonds, property values, quarterly net profit after tax, securities' closing prices, book value (BE), market value (ME), daily VN-index, and treasury bill yields is sourced from the State Bank of Vietnam's official website.
In this analysis, I utilize data from phase 1 to evaluate the return costs of individual stocks, the market fund's return rate, and key financial ratios including the Book-to-Market (B/M) ratio, market capitalization (Size), net profit after tax (OP), total asset growth (Investment pattern), and the risk-free interest rate based on Vietnamese Treasury bills.
Step 3: Dividing and building up portfolios
According to the 4 quotas, including scale, B/M ratio, OP, and Inv divided yield to 18 portfolios will be created, and the detailed information will be given in the following section
This article explores five key financial variables in Microsoft Excel: Small minus Big (SMB), which measures the return difference between small and large stocks; High minus Low (HML), indicating the return disparity between high and low book-to-market ratio stocks; Robust minus Weak (RMW), reflecting the return variation between stocks with strong and weak profitability; and Conservative minus Aggressive (CMA), which assesses the return difference between conservative and aggressive investments The article details the statistical calculations for these variables, examines their interrelationships, and employs regression methods to analyze their interactions.
Step 5: Running the simulations, then doing the regression study
Efficiently managing variable confusion in Excel portfolios using Stata involves analyzing the return variance of explanatory variables, conducting regression models, and assessing the practical utility of these models for specific business applications.
Step 6: Analysis of research results and give conclusions
The findings indicate that specific causes significantly influence investment rates, leading to a deeper exploration of mathematical regression research to inform final decisions for investors and business owners.
Data is analyzed using Stata version 13 and regression results are generated Microsoft Office Excel is used to incorporate the sample results.
Model
The model for time-series research:
(4) is the expected return on porfolio i for period t is the risk-free interest rate of government bonds for period t is the excess market return for period t
The Small Minus Big (SMB) factor introduced by Fama and French in 1993 represents the difference in average returns between small-cap and large-cap stocks within an industry This factor is calculated using portfolios based on market capitalization, which serves as a proxy for size, derived from financial statements for the quarter ending in t -1.
The High Minus Low (HML) risk factor, introduced by Fama and French in 1993, measures the return disparity between portfolios of stocks with the highest and lowest book-to-market ratios This metric highlights the performance gap on day t, based on the inventory values to revenue ratios from the preceding fiscal quarter, t-1.
The Robust Minus Weak Low (RMW) is a risk factor introduced by Fama and French in 2015, highlighting the average performance disparity between top and bottom-performing stocks within an industry To assess profitability, portfolios are constructed quarterly, utilizing accounting results from the previous period (t -1).
The Conservative Minus Aggressive (CMA) risk factor, introduced by Fama and French in 2015, highlights the performance disparity between the most cautious and the most aggressive portfolios in the capital market These portfolios are constructed quarterly, based on the increase in net assets from the previous fiscal quarter, divided by the total assets at the end of that quarter.
The Fama and French strategy, established in 1993 and updated in 2015, involves a systematic approach to portfolio allocation based on four key factors: market capitalization, book-to-market ratio, profitability, and investment Each quarter, firms are sorted and categorized into portfolios according to these criteria, facilitating a structured investment process.
The stock return represents the average rate of return calculated over the days within a quarter To determine this, take the ending price of the stock at quarter \( t \) and compare it to the ending price at quarter \( t-1 \) This allows for an approximation of the stock's rate of return for the specified quarter.
The market return is derived from the VN-Index on a daily basis and is assessed quarterly, influenced by the interest rates of each quarter The VN-Index for quarter t is specifically referred to as VN-Index quarter t, and the regular rate of return is calculated based on this index.
The risk-free rate of return, representing the "Market," is defined as the principal interest rate for a 1-year Treasury bill issued by the Reserve Bank of Vietnam from January 2014 to December 2019 This rate is calculated by dividing the quarterly risk-free interest rate by four quarters.
Market capitalization (stands for “Size”): the product of number of shares outstanding and the market price per share, as on the last day of each quarter t:
Profitability, often referred to as "OP," is defined as the return on equity (ROE) It is calculated by subtracting all operating expenses, interest, depreciation, taxes, and preferred stock dividends from a company's total revenue, resulting in the net sales that reflect the company's financial performance.
Investment (stands for “Inv”): using asset growth as a proxy for investment If the total assets in quarter t and the total assets in quarter t-1, following Cooper et al
(2008), Fama and French (2008), Gray and Johnson (2011) and Fama and French
(2014) Asset growth is defined as follows:
Starting in January 2014, the study by Brailsford et al analyzes judgment across cut-off portfolios, focusing on the sorting and construction of portfolios The research employs factors similar to the methodology outlined by Fama and French (2015), utilizing a 2x3 sort, while other sorting methods have been applied with limited guiding principles This lack of consensus on suitable sorting approaches leads to variations in their effectiveness across different datasets Consequently, the study aims to sort 100 major business stocks based on expenditure, scale, and profitability.
First, to compiling SMB, firms are divided into two groups: Small (S) and Big (B) – on the basis of market capitalization (using median value as the break point)
To define HML, companies are categorized into three groups based on their Book-to-Market (B/M) ratio: High (H), Neutral (N), and Low (L) This classification results in three portfolios determined by breakpoints for the bottom 40%, middle 20%, and top 40% of values The 2x3 sorting process generates six distinct portfolio blocks, as illustrated in Table 1.
Table 1 Size and B/M bivariate sorting
The process for identifying RMW (Robust Minus Weak) and CMA (Conservative Minus Aggressive) follows the same methodology, differing only in the second sorting based on specific variables: Robust (R), Neutral (N), and Weak (W) for RMW, and Conservative (C), Neutral (N), and Vigorous (A) for CMA The outcomes of these alternate portfolios are illustrated in Tables 2 and 3.
Table 2 Size and Investment bivariate sorting
Table 3 Size and Profitability bivariate sorting
This analysis examined the return on investment (ROI) associated with the Fama-French five-factor model We utilized the quarterly returns of 18 portfolios, incorporating size, market equity to book equity (ME/BE), operating profitability (OP), and investment (Inv) as dependent variables Additionally, we employed excess returns from mimicking portfolios, including Small Minus Big (SMB), High Minus Low (HML), Robust Minus Weak (RMW), and Conservative Minus Aggressive (CMA), as explanatory variables in our regression analysis.
Factors calculating
From the model developed by Fama and French, we can see that there are three explanatory variables involved
SMB (SH+SN+SL+SR+SN+SW+SC+SN+SA)/9 –
(BH+BN+BL+BR+BN+BW+BC+BN+BA)/9
HML (SH+BH)/2 – (SL+BL)/2
RMW (SR+BR)/2 – (SW+BW)/2
CMA (SC+BC)/2 – (SA+BA)/2
Table 4 Construction of size, ME/BE, profitability and investment factors
The six portfolios categorized by Size and Book-to-Market (B/M) ratios are labeled as SH, SN, SL, BH, BN, and SMB The SMB metric is calculated as the difference between the total of small size portfolios and the total of big size portfolios, represented by the formula: [(SH + SN + SL + SR + SN + SW + SC + SN + SA) / 9].
The portfolio value coefficient (HML) is calculated by excluding neutral portfolios, represented as [(SH+BH)/2 - (SL+BL)/2] When evaluating the profitability and investment shares of the 2x3 sorts, particularly RMW and CMA, it is important to consider using operating profit (OP) and investment (Inv) instead of book-to-market (B/M) as the secondary sorting criterion The author's initial findings were derived through a step-by-step experimental process By employing multiple regression analysis, the explanatory variables are ranked based on their marginal contributions, determining their significance in the regression model This analysis also assesses whether any explanatory variables can be removed due to redundancy once other variables are included The model selection process is finalized when no further explanatory variables can be added or excluded from the regression equation.
Testing methods and Hypotheses of research
First, use Ordinary Least Square (OLS) test applied to Fama Macbeth two-way Regression
This study employs time-series regression to examine the relationship between five risk factors and market returns for each portfolio Utilizing Ordinary Least Squares (OLS) models, the analysis focuses on describing these risk factors in financial markets The Betas derived from the model are calculated and tested to ensure their significance, following the methodology established by Fama and French (1992).
In their 1992 study, Fama and French established that a high exchange risk often correlates with a greater likelihood of returns, highlighting the strong relationship between risk factors such as SMB, HML, RMW, and CMA Their 2015 findings suggest that these factors are positively interconnected, indicating that investors can expect rewards for embracing the risks tied to their investments However, in the Vietnamese stock market, investors tend to deviate from standard theories, yet research continues to yield positive outcomes across various factors.
: There is a positive correlation between the market factor and the excess return of the portfolio
: There is a positive correlation between the SMB factor and the excess return of portfolio
: There is a positive correlation between the HML value factor and the excess return portfolio
: There is a positive correlation between the RMW profit factor and the return on investment return
: There is a positive correlation between the CMA profit factor and the return on investment return
Heteroskedasticity White test is applied
The White test for homoskedasticity is a test to ensure that the errors in a regression model are normally distributed
Hypothesis : There is no hetoroskedasticity
If p-value≤0 05: reject , There is hetoroskedasticity
If p-value>0.05: accept , There is no hetoroskedasticity
Finally, use the GRS Regression test
The GRS test, established by Gibbons et al in 1989, evaluates the mean-variance performance comparison between a left-hand-side array of assets or portfolios and a right-hand-side model or portfolio.
N and T – N – K degrees of freedom (assumed that the errors are homoskedastic and uncorrelated)
T is the total number of observations
N is the number of assets (in this case, stocks)
K is the number of factors
The sample mean of the factor returns is denoted as (f), while the sample variance matrix of these returns is represented as ̂ The estimated alphas from the multivariate regression are indicated, and ̂ signifies the covariance matrix of the residuals from this regression analysis.
The GRS test evaluates the significance of alpha qualities in individual model regressions to determine if a model accurately captures sample return variance A GRS score of zero for each regression indicates that the overall GRS score will also be zero Higher GRS statistic values reflect larger alphas, moving further from zero, indicating poorer performance of the asset-pricing model.
The regression coefficient plays a crucial role in financial literature, supported by various statistical tests Among these, the GRS-F test proposed by Gibbons et al (1989) is essential for assessing whether the alpha coefficients of data sets are significantly different from zero Due to space limitations, the significance level, or p-value, for the GRS-F calculation is not detailed here (Gibbons et al., 1989, p 1124).
: All coefficients of Fama French five-factor got from various variables are identical to zero ( =0)
: Not all coefficients of Fama French five-factor got from various variables are not identical to zero ( ≠0)
Chapter 3 utilizes a comprehensive study of multiple database references and simulation methodologies to explore key indicators of major asset pricing models, including measures of firm returns, price risk, size risk, value risk, profitability risk, and investment pattern risk This section will also elucidate the regression analysis tool and the interpretation of the results.
EMPERICAL RESULTS
Descriptive statistics
Table 5 Stationarity test results regarding level values of variables
Deviation Minimum Maximum Skewness Kurtosis
Source: data collected by the author and calculated on Stata version 13 software
Table 5 Panel A presents the descriptive statistics for the six-year intersection portfolio, highlighting the quarterly factor premiums in industry companies The quarterly returns for the factor portfolios are recorded as 0.043% for MRP, -0.016% for SMB, -0.060% for HML, 0.034% for RMW, and -0.013% for CMAs When these portfolio returns are arranged from highest to lowest, they reflect the varying performance of the factors.
Recent estimates indicate that the excess return on risk-free markets typically surpasses the excess returns of various portfolio strategies, including robust versus low-income, conservative versus aggressive, small versus big, and high versus low book-to-market (B/M) ratios Notably, the standard deviation of these results is relatively low, suggesting that the outcomes are closely grouped around the mean The smallest deviation observed is 0.0565 for the small-minus-big (SMB) portfolios and 0.0756 for the market risk premium (MRP) portfolios.
MRP SMB HML RMW CMA
Correlation analysis provides an overview of the relationships among study variables, highlighting how one variable reacts to changes in another The linear regression coefficient indicates whether a relationship exists between two variables Notably, there is a negative relationship between the market portfolio and CMW, while a weak positive relationship exists between the market portfolio and SMB The correlation coefficient between the market portfolio and the HML factor exceeds 63.39%, indicating a significant link where corporate value shifts impact the stock market Additionally, the HML factor positively correlates with the SMB factor, suggesting that larger listed enterprises balance their importance Conversely, the RMW factor exhibits the weakest correlations with other factors, particularly with SMB, indicating minimal impact from industrial business capital The CMA factor shows positive connections with both the demand factor and SMB, suggesting that a vibrant market enhances investment in listed industries and increases their market capital Overall, the correlation factors among the explainable variables are generally weak, yet they encompass returns from the research companies and the Fama French five-factor model.
The author presents a table illustrating the association measures among variables, highlighting that repetitive knowledge can distort regression effects The hypothesis states that multicollinearity is indicated by a tolerance value below 0.2 or 0.1, alongside a VIF of 10 or higher, which the author supports based on the findings Ultimately, the secondary regression model is confirmed, demonstrating no multicollinearity among the variables The VIF analysis reveals a maximum VIF of 1.87 and a mean VIF of 1.49, indicating that multicollinearity does not impact the regression results.
Regression details
The analysis categorizes portfolios into various types: small and big (S//B), medium (M), high and low (H//L), robust and weak (R//W), and conservative and aggressive (C//A) The significance of the coefficients is indicated by t statistics in parentheses and P-values in brackets, with levels of significance marked as follows: (***) for 1%, (**) for 5%, and (*) for 10% Additionally, the t statistics have been adjusted using the Newey-West method to address heteroscedasticity concerns.
The market risk factor ( ) – MRP
The market risk premium coefficients for both fund managements are significantly optimistic, with the exception of the RMW-BN Additionally, the MRP factor coefficients play a substantial role in determining the average returns of these portfolios.
0,391 (RMW-BN portfolio) to 1,321 (SMB-BH portfolio) These findings show that
MRP does not have a major impact on the performance of the large scale firms, and is consistent with large gains and strong B/M ratios
The size risk factor ( ) – SMB
In 11 out of 18 portfolios, the size factor is significant, with substantial levels at 1% and 5%, while the CMA-BN portfolio shows significance at the 10% level The SMB-BN coefficients range from a low of -1.286 to a high of 0.415 for the SMB-SH portfolio Notably, the size factor correlates with average yields for SMB-SH, RMW-SR, and major firms, including the CMA trend factor Overall, the SMB element displays a lack of negative correlation with large industrial firms, often being more pronounced in smaller businesses.
The value risk factor ( ) – HML
Among the 18 portfolios analyzed, five demonstrate significant importance at both the 1 percent and 5 percent significance levels, while the remaining 13 portfolios are deemed negligible at these thresholds The HML component exhibited a negative response to SMB-SH, BL, CMA-SC, SN, SA, BC, and BA, with coefficients ranging from -0.648 to -0.091, whereas it showed a positive reaction to SMB-SN.
SL, BH, BL, all RMW portfolios, and all the CMA-BN with coefficients ranging from 0.190 to 0.807
The profitability risk factor ( ) – RMW
The findings indicate that 33% of the total portfolios exhibit significant performance at levels of 1%, 5%, and 10% Additionally, other portfolios demonstrate favorable responses with coefficients ranging from 0.104 to 0.582, reflecting the negative reactions of RMW-SR, SN, and SW.
The findings related to the BN and BW portfolios align with the model's predictions, indicating that low-benefit portfolios exhibit a negative return on investment (r i), while high-productivity portfolios show more favorable outcomes, albeit with some exceptions This trend enhances the efficiency of the Association and contributes to a more consistent inventory management.
The investment risk factor ( ) – CMA
At the 1% and 5% significance levels, the CMA factor is theoretically significant for CMA-SC, SN, and BA portfolios Notably, industrial firms with aggressive investments, such as SA and BA, provide a compelling insight into the investment factor.
77 percent and 86 percent projected returns Based upon the portfolios of low investment-trend, low coefficients are shown to concentrate on the portfolios SMB-
SH, SL, BN, BL, RMW-SN, CMA-SN, SA, BA and vice versa As the company builds the company, its usual stock drops.
Relevant test
The White test for Heteroskedasticity test
In 1980, White conducted a heteroskedasticity test by regressing the squared residuals against all unique regression variables, their cross-products, and squares The resulting p-value, distributed as a Chi-squared statistic under the null hypothesis of homoskedasticity, was found to be 0.9091, which is greater than the significance level of 10% Therefore, the null hypothesis (H0) is accepted, indicating that the model does not exhibit heteroskedasticity.
In an asset price model, the alpha intercept estimates should be nearly zero to ensure the model's effectiveness in evaluating excess returns over risk-free rates The following test will provide insights into the influence of various factors on the overall returns of investment portfolios.
Dependent variables Model Average GRS
SMB_SH, SMB_SN, SMB_SL, SMB_BH,
SMB_BN, SMB_BL, RMW_SR, RMW_SN,
RMW_SW, RMW_BR, RMW_BN,
RMW_BW, CMA_SC, CMA_SN, CMA_SA,
CMA_BC, CMA_BN, CMA_BA
This table presents the excess return rates from 18 portfolios categorized by size, valuation, profitability, and investment It includes average values, GRS-F test statistics, and P-values for the analysis A notable finding is that the average percentage of 0.6581 is highlighted, indicating the model's effectiveness Clearly, the Fama-French five-factor model demonstrates significant capability in explaining variations in excess portfolio returns.
The GRS-F test result for the Fama French five-factor model is 1.21, with a P-value of 0.48, leading to the acceptance of the null hypothesis This indicates that the model effectively explains excess returns without the need for additional factors.
About the result
The study reveals that variables within the sequence hold a significant rating of up to 90.77%, with the business risk premium exerting the most substantial influence on manufacturing firms' returns A total of 18 portfolios effectively align with a random vector of elements, while SMB demonstrates the second highest explanatory power for fund returns The analysis indicates that the size factor negatively impacts company return directions, and large portfolios show no correlation with the RMW component Despite this, it cannot be concluded that overconfidence (OP) does not affect market prices, as OP significantly influences consumer perceptions and the strategies of large corporations with high-margin BR portfolios Investors remain concerned about firm capitalization and future organizational performance Additionally, the application of HML to SMB highlights the importance of coefficients and marginal costs, while the CMA effect, though less critical, remains statistically relevant across all regression coefficients.
This section explores the relationships between excess returns in the Fama-French five-factor model and manufacturing firms The author employs descriptive statistics and database analysis to validate the explanatory model To enhance prediction accuracy, necessary tests are performed to mitigate common issues like multicollinearity and heteroskedasticity The regression results will be analyzed in alignment with contemporary scientific data.
CONCLUSION AND RECOMMENDATIONS
Conclusion
The aim of this analysis is to apply the Fama French five-factor asset pricing model to
This study examines the relationship between excess returns of portfolios and five factors—market, size, value, profitability, and investment—using data from 100 companies listed on stock exchanges from January 2014 to December 2019 The findings reveal that all five characteristics are statistically correlated to excess returns across various portfolio sorts, including Size-B/M, Size-OP, and Size-Inv While the CMA factor supports the other four, its impact is minimal, suggesting it may be redundant in a streamlined model The remaining three factors demonstrate significant power, but their effects can obscure each other when used in portfolio creation For instance, the size factor shows that smaller stocks tend to outperform larger ones, while the value factor indicates that stocks with high B/M ratios gain more than those with low ratios Notably, the market factor significantly influences larger companies, driving stock prices upward in a growing market, while high B/M stocks tend to benefit more during market upswings Conversely, low B/M stocks may suffer indirectly due to the attractiveness of high B/M stocks The analysis supports the validity of the five-factor model, as confirmed by average OLS regression intercepts and GRS tests, which collectively endorse the model's capacity to describe average returns across all tested portfolios.
Recommendations
5.2.1 Recommendations for those who use the Fama French five-factor model
The FF model fails to predict significant fluctuations in equity markets during extreme volatility, as evidenced by the 2007 bubble reports in Vietnam and the drastic market price drop in 2008 due to the global economic downturn Consequently, the crisis phases can be excluded when applying FF templates This analysis serves as a valuable resource for students and researchers utilizing the model to align their comparisons and objectives effectively.
Demand conditions significantly influence the economy, making them a crucial factor in selecting the multi-factor model for analysis A simpler approach, utilizing a four-factor model comprising 𝑟, SMB, HML, and RMW, can be effective The Fama-French five-factor model is comparable to other common models in predicting stock price outcomes However, certain stocks may not conform to these effects, resulting in negative coefficients for some portfolios when all factors are considered This regression phenomenon, where a specific factor exhibits negative results for a limited number of categories, is not exclusive to Fama-French analysis but can be observed across various securities and industries.
When investing in stocks, it is essential for investors to consider factors such as interest, scale, volume, and potential benefits, as these significantly influence market fluctuations Understanding how these elements impact stock valuation is crucial, especially since market conditions can vary based on the size of the stocks Larger organizations tend to experience greater profits and losses, prompting investors to evaluate evolving markets carefully Conversely, in uncertain sectors, it is advisable for investors to opt for stocks with lower book-to-market ratios and share prices to mitigate risks while still seeking beneficial returns.
This investment model highlights three key factors: scale, valuation, and benefit When the return gap between small-cap stocks (SMB) and large-cap stocks widens, investors should consider selling large-cap stocks to invest in small-cap stocks Additionally, it is advisable to sell stocks with low book-to-market (B/M) ratios and invest in those with strong B/M ratios for better returns.
Investors often prioritize scale factors after business considerations, distinguishing between returns from small and large stocks By calculating the Quality Capital Indicator (QCI), they can identify opportunities to buy smaller securities while divesting from larger ones This strategy allows buyers to minimize risk while maximizing profits by purchasing low-risk stocks and selling those with higher returns.
Investors should monitor the average returns from strong operational performance (OP) and robust OP If the gap between these returns is decreasing, it may be wise to sell weak-OP stocks and invest in extremely strong-OP stocks Additionally, investors must remain cautious about corporate claims regarding earnings, as these figures could be exaggerated or misleading.
The Fama-French hypothesis suggests that while more stocks attract buyers, it overlooks the impact of real stocks on public perception Poor company performance can lead to decreased stock prices and increased selling activity Investors should not feel compelled to buy securities with expected decent results to achieve similar returns In fact, shares of non-profit firms with low stock valuations may still provide substantial future returns.
5.2.3 Recommendations for stock market in Vietnam
Large businesses with a high book-to-market (B/M) ratio face challenges due to changes in the business climate, unpredictable investment patterns, and declining earnings To mitigate these risks, companies should closely manage their stock and property scales while minimizing reliance on the stock market Conversely, if businesses believe that the economic environment will improve, they may choose to downsize, which could lead to an increase in their B/M ratio.
A low operating margin can negatively affect businesses by leading to reduced revenues, but successful companies can mitigate various risks, especially market and valuation risks In times of contraction, overall performance may decline, and a low book-to-market (B/M) ratio on the balance sheet can further result in diminished operating profits.
This research has notable limitations, including a data range limited to six years and a focus primarily on the industrial sector, which restricts its ability to represent the entire market and accurately reflect the impact of various factors on average stock returns Consequently, both the size and duration of the database are relatively small Methodologically, while Fama and French (2014) suggest multiple portfolio division methods, this study employs only a 2x3 sort factor approach and does not utilize the full two-way Fama French regression, opting instead for the GRS F-test to assess the model's effectiveness.
Further research on pricing formulas could enhance the understanding of small and medium-sized businesses (SMBs) by organizing portfolios based on various factors Future studies may focus on the evaluation of business portfolios across different market capitalizations, such as small-cap, mid-cap, and large-cap It remains essential to investigate whether this segregation truly exists Fama and French (2016) noted that cash profitability outperforms operational profitability within a five-factor model, suggesting that testing this outcome in Vietnam's markets could reveal insights into the causes of productivity losses Acknowledging the imperfections of the five-factor model is crucial, and findings from backward calculations indicate that additional elements, such as Return on Assets (ROA) and Earnings Before Interest and Taxes (EBIT), may also impact stock prices on the Vietnam stock exchange.
1 Trương Đông Lộc and Dương Thị Hoàng Trang (2014), “Mô hình 3 nhân tố Fama-French: Các bằng chứng thực nghiệm từ Sở giao dịch chứng khoán Thành phố Hồ Chí Minh”, Cần Thơ University Science Magazine
2 Võ Hồng Đức and Mai Duy Tân (2014), “Sự phù hợp của mô hình Fama- French 5 nhân tố cho thị trường chứng khoán Việt Nam”, Technology banking magazine, no 101, pages 2-20
3 Nguy n Thị Thúy Nhi (2016), “Kiểm định mô hình Fama-French 5 nhân tố và mô hình Q bốn nhân tố trên thị trường chứng khoán Việt Nam” Open University HCMC
4 Huỳnh Ngọc Minh Trâm (2017), “V n dụng mô hình Fama French 5 nhân tố để ước lượng tỷ suất lợi tức kỳ vọng của các cổ phiếu niêm yết trên sàn giao dịch chứng khoán Hồ Chí Minh”, Da Nang Economic University
5 Ross S (1976), "The Arbitrage theory of capital asset pricing", Journal of Economic Theory
6 Nartea et al (2009), “Extreme returns in emerging stock markets: evidence of a MAX effect in South Korea”
7 Merton H Miller and Franco Modigliani (1961), “Dividend Policy, Growth, and the Valuation of Shares”
8 Michael C Jensen, Fisher Black (1972), “The Capital Asset Pricing Model: Some Empirical Tests”
9 Hou et al (2012, 2015), “An Empirical Assessment of the Q-Factor Model”, The Lahore Journal of Economics
10 Robert Novy-Marx (2012), “The Other Side of Value: The Gross Profitability Premium”
11 Carthart, M (1997), “On persistence of Mutual Fund Performance”, The Journal of Finance
12 John H Cochrane (2011), “Presidential Address: Discount Rates”
13 Narasimhan Jegadeesh, Sheridan Titman (2001); “Profitability of Momentum Strategies: An Evaluation of Alternative Explanations”
14 Fama, Eugene F.; MacBeth, James D (1973), “Risk, Return, and Equilibrium: Empirical Tests”
15 JH Cochrane (2011), “Presidential address: Discount rates”, The Journal of Finance
16 Eugene F Fama, Kenneth R French (2014), “A five-factor asset pricing model”, Journal of Finance
17 Souad Ajili (2002), “Capital Asset Pricing Model and three factor model of Fama and French revisited in the case of France”
18 John M Griffin (2002), “Are the Fama and French Factors Global or Country- Specific?”
19 Keiichi Kubota and Hitoshi Takehara (2017), “Does the Fama and French Five‐ Factor Model work well in Japan?”
20 Grace Xing Hu, Can Chen, Yuan Shao, Jiang Wang (2018); “Fama–French in China: Size and Value factors in Chinese Stock Returns”
21 Harshita, S Singh, Surendra S Yadav (2015), “Indian Stock Market and the Asset Pricing Models”
22 Songül Kakilli Acaravci, Yunus Karaomer (2017), “Fama-French five factor model: Evidence from Turkey”
23 Ferson, W E , and C R Harvey (1999), “Conditioning variables and the cross- section of stock returns,” Journal of Finance
24 Kent Daniel, Sheridan Titman, and K C John Wei (2001); “Explaining the Cross-Section of Stock Returns in Japan: Factors or Characteristics?”, Journal of Finance
25 Heston, S L , Rouwenhorst, K G , & Wessels, R E (1999), “The Role of Beta and Size in the Cross-Section of European Stock Returns”, European Financial Management
26 Gibbons, M R , Ross, S A , Shanken J (1989); “A test of the efficiency of a given portfolio”, Econometrica
Make OLS Regression test with 18 sorted portfolios:
Regression of SMB- SN with Fama French five-factor
Regression of SMB- BH with Fama French five-factor
Regression of CMA- BA with Fama French five-factor
Regression of CMA- BC with Fama French five-factor
Regression of CMA-BN with Fama French five-factor
Regression of CMA-SA with Fama French five-factor
Regression of CMA-SC with Fama French five-factor
Regression of CMA-SN with Fama French five-factor
Regression of RMW-BN with Fama French five-factor
Regression of RMW-BR with Fama French five-factor
Regression of RMW-BW with Fama French five-factor
Regression of RMW-SN with Fama French five-factor
Regression of RMW- BR with Fama French five-factor
Regression of RMW- SW with Fama French five-factor
Regression of SMB- BL with Fama French five-factor
Regression of SMB- BN with Fama French five-factor
Regression of SMB- SH with Fama French five-factor
Regression of SMB- SL with Fama French five-factor