Fama French factor returns

We compile all return and macroeconomic data from Kenneth French's website and the OECD statistical data warehouse, respectively, for the period from January 1990 to December 2018. All return and macroeconomic data include the following countries: Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Netherlands, Norway, Portugal, Spain, Sweden, Switzerland and United Kingdom.The dataset comprises the following series:

1. Fama-French factors, 3-factor model, as provided by Kenneth French (Europe_3_Factors.txt).
2. Fama-French factors, 5-factor model, as provided by Kenneth French (Europe_5_Factors.txt).
3. Returns for 25 size-BE/ME portfolios, as provided by Kenneth French (Europe_25_Portfolios_ME_BE-ME.txt).
4. Returns for 25 size-momentum, as provided by Kenneth French (Europe_25_Portfolios_ME_Prior_12_2.txt).
5. Weighted average per capita consumption growth. We first collect quarterly chained volume estimates for consumption in nondurables and services, non-seasonally adjusted, in national currency, for the 16 countries under consideration (‘Non-durable goods’ and ‘Services’ series, LNBQR measure). Second, we use the population series provided by the OECD to determine per capita consumption growth series for each country. Finally, we estimate the average consumption growth for the economies under consideration, weighting by population (Europe_Consumption_Q.txt).
6. Weighted average consumer confidence index (CCI). We collect monthly CCI data as provided by the OECD (‘OECD Standardised CCI, Amplitude adjusted, sa’ series, dataset ‘Composite Leading Indicators’, MEI). We determine the average CCI for the economies under consideration, weighting by population (Europe_Indicators_Q.txt).

Subscribers can find out export and import data of 23 countries by HS code or product’s name. This demo is helpful for market analysis.

ABSTRACT The purpose of this work is to present the Weighted Forward Search (FSW) method for the detection of outliers in asset pricing data. This new estimator, which is based on an algorithm that downweights the most anomalous observations of the dataset, is tested using both simulated and empirical asset pricing data. The impact of outliers on the estimation of asset pricing models is assessed under different scenarios, and the results are evaluated with associated statistical tests based on this new approach. Our proposal generates an alternative procedure for robust estimation of portfolio betas, allowing for the comparison between concurrent asset pricing models. The algorithm, which is both efficient and robust to outliers, is used to provide robust estimates of the models’ parameters in a comparison with traditional econometric estimation methods usually used in the literature. In particular, the precision of the alphas is highly increased when the Forward Search (FS) method is used. We use Monte Carlo simulations, and also the well-known dataset of equity factor returns provided by Prof. Kenneth French, consisting of the 25 Fama-French portfolios on the United States of America equity market using single and three-factor models, on monthly and annual basis. Our results indicate that the marginal rejection of the Fama-French three-factor model is influenced by the presence of outliers in the portfolios, when using monthly returns. In annual data, the use of robust methods increases the rejection level of null alphas in the Capital Asset Pricing Model (CAPM) and the Fama-French three-factor model, with more efficient estimates in the absence of outliers and consistent alphas when outliers are present.

This paper illustrates the pitfalls of the conventional heteroskedasticity and autocorrelation robust (HAR) Wald test and the advantages of new HAR tests developed by Kiefer and Vogelsang in 2005 and by Phillips, Sun and Jin in 2003 and 2006. The illustrations use the 1993 Fama-French three-factor model. The null that the intercepts are zero is tested for 5-year, 10-year and longer sub-periods. The conventional HAR test with asymptotic P-values rejects the null for most 5-year and 10-year sub-periods. By contrast, the null is not rejected by the new HAR tests. This conflict is explained by showing that inferences based on the conventional HAR test are misleading for the sample sizes used in this application.

Using all stocks listed in the Tokyo Stock Exchange and macroeconomic data for Japan, the dataset comprises the following series:

1. Monthly returns for 25 size-book-to-market equity portfolios, following the Fama and French (1993) methodology. (Raw data source: Datastream database)
2. Monthly returns for 20 momentum portfolios, following the Fama and French (1993) methodology. (Raw data source: Datastream database)
3. Monthly returns for 25 price-to-cash flow-dividend yield portfolios, following the Fama and French (1993) methodology. (Raw data source: Datastream database)
4. Fama and French three-factors (RM, SMB and HML), following the Fama and French (1993) methodology. (Raw data source: Datastream database)
5. Fama and French five-factors (RM, SMB, HML, RMW, and CMA), following the Fama and French (2015) methodology for all factors, except for RMW, which is determined using the return on assets as sorting variable, as in Hou, Xue and Zhang (2014). (Raw data source: Datastream database)
6. Private final consumption expenditure, in national currency and constant prices, non-seasonally adjusted, for Japan. (Raw data source: OECD)
7. Consumer Confidence Index (CCI) for Japan. (Raw data source: OECD)
8. Three-month interest rate of the Treasury Bill for Japan. (Raw data source: OECD)
9. Gross Domestic Product (GDP) for Japan. (Raw data source: OECD)
10. Consumer Price Index (CPI) growth rate for Japan. (Raw data source: OECD)

We have produced all return series using the following data from Datastream: (i) total return index (RI series), (ii) market value (MV series), (iii) market-to-book equity (PTBV series), (iv) total assets (WC02999 series), (v) return on equity (WC08301 series), (vi) price-to-cash flow ratio (PC series), and (vii) dividend yield (DY series). We have used the generic rules suggested by Griffin, Kelly, & Nardari (2010) for excluding non-common equity securities from Datastream data. We also exclude stocks with less than twelve observations in the period from July 1992 to June 2018. Accordingly, our sample comprises a total number of 5,312 stocks.

REFERENCES:

Fama, E. F. and French, K. R. (1993). Common risk factors in the returns on stocks and bonds. Journal of Financial Economics, 33, 3–56. Fama, E. F. and French, K. R. (2015). A five-factor asset pricing model. Journal of Financial Economics, 116, 1–22. Griffin, J. M., Kelly, P., and Nardari, F. (2010). Do market efficiency measures yield correct inferences? A comparison of developed and emerging markets. Review of Financial Studies, 23, 3225–3277. Hou K, Xue C, Zhang L. (2014). Digesting anomalies: An investment approach. Review of Financial Studies, 28, 650-705.

We compile raw data from the Datastream database for all stocks traded on the Spanish equity market. Particularly, we compile the following data series: (i) total return index (RI series), (ii) market value (MV series), (iii) market-to-book equity (PTBV series), (iv) total assets (WC02999 series), (v) return on equity (WC08301 series), (vi) dividend yield (DY series), (vii) price-to-earnings ratio (PE series), and (viii) effective tax rate (WC08346 series). We use the filters suggested by Griffin, Kelly, and Nardari (2010) for the Datastream database to exclude assets other than ordinary shares from our sample. Hence, our sample comprises 443 companies, including all firms that started trading within the time interval under study, as well as those that were delisted. As a proxy for the risk-free rate, we use the three-month Treasury Bill rate for Spain, as provided by the OECD. Accordingly, the dataset comprises the following series:

1. Spain_9_Portfolios_SIZE_BEME: Monthly returns for 9 size-book-to-market equity portfolios, following the Fama and French (1993) methodology. (Raw data source: Datastream database)
2. Spain_9_Portfolios_DY_PE: Monthly returns for 9 dividend yield-price-to-earnings ratio, following the Fama and French (1993) methodology. (Raw data source: Datastream database)
3. Spain_9_Portfolios_SIZE_TR: Monthly returns for 9 size-effective tax rate portfolios, following the Fama and French (1993) methodology. (Raw data source: Datastream database)
4. Spain_FF_3_Factors: Monthly returns for the constituents of the three classic factors of Fama and French, following the Fama and French (1993) methodology. (Raw data source: Datastream database)
5. Spain_FF_5_Factors: Monthly returns for the constituents of the five factors of Fama and French, following the Fama and French (2015) methodology. (Raw data source: Datastream database)
6. Spain_RF: Three-month Treasury Bill rate for Spain. (Raw data source: OECD)
7. Spain_Avg_Tax_Rate: Value-weighted effective tax rate paid by companies traded in Spain. (Raw data source: Datastream database)

REFERENCES:

Fama, E. F. and French, K. R. (1993). Common risk factors in the returns on stocks and bonds. Journal of Financial Economics, 33, 3–56. Fama, E. F. and French, K. R. (2015). A five-factor asset pricing model. Journal of Financial Economics, 116, 1–22. Griffin, J. M., Kelly, P., and Nardari, F. (2010). Do market efficiency measures yield correct inferences? A comparison of developed and emerging markets. Review of Financial Studies, 23, 3225–3277.

Using an intangible intensity factor that is orthogonal to the Fama–French factors, we compare the role of intangible investment in predicting stock returns over the periods 1963–1992 and 1993–2022. For 1963–1992, intangible investment is weak in predicting stock returns, but for 1993–2022, the predictive power of intangible investment becomes very strong. Intangible investment has a significant impact not only on the MTB ratio (Fama-French HML factor) but also on operating profitability (Fama-French RMW factor) when forecasting stock returns from 1993 to 2022. For intangible asset-intensive firms, intangible investment is the main predictor of stock returns, rather than MTB ratio and profitability. Our evidence suggests that intangible investment has become an important factor in explaining stock returns over time, independent of other factors such as profitability and MTB ratio.

The replication data contain MATLAB and GAUSS codes as well as the data required for replication of the results from the paper

1. Monte Carlo Simulation:

Contains codes and data for simulation study from Section 3.

Data:

• MV.mat, MV.txt- monthly data on market capitalization of the 205 stocks of the S&P500 index obtained from DataStream for the period 01.01.1974-01.05.2015

• sp500_edata.mat - monthly data on close prices of components of S&P500 index for the period 01.01.1974-01.05.2015 processed to obtain excess returns using as a risk free return data on the risk free return from French & Fama database. Description of the price data from DataStream: "The ‘current’ prices taken at the close of market are stored each day. These stored prices are adjusted for subsequent capital actions, and this adjusted figure then becomes the default price offered on all Research programs. " Description of the excess return of the market from French & Fama database : "the excess return on the market, value-weight return of all CRSP firms incorporated in the US and listed on the NYSE, AMEX, or NASDAQ that have a CRSP share code of 10 or 11 at the beginning of month t, good shares and price data at the beginning of t, and good return data for t minus theone-month Treasury bill rate (from Ibbotson Associates)." From the latest file two separate data files were created (see CAPMsim.m):

• sp500_stocks.txt, sp500_stocks.mat - monthly data on close prices of 205 components of S&P500 index for the period 01.01.1974-01.05.2015

• FactorData.txt, FactorData.txt - The Fama & French factors from French & Fama database for a period July 1926 - May 2015.

Codes:

• CAPMsim.m - the main code that replicates the Monte Carlo simulation of the artificial market and proxy indexes subject to different types of the measurement error.

• sure.m- obtains the estimated parameters for the SUR system and performs hypothesis testing of the significance of the coefficients.

2. Empirical Application

Contains codes and data for empirical application from Section 4.

Data:

• data1203.txt - 120 monthly observations on the excess returns on 20 random stocks from S&P500, S&P500 index return, DJIA return from DataStream and excess return of the CRSP index from French & Fama database for a period 01/06/2005-01/05/2015.

• data1204.txt - 120 monthly observations on the excess returns on 30 stocks from DJIA, S&P500 index return, DJIA return from DataStream and excess return of the CRSP index from French & Fama database for a period 01/06/2005-01/05/2015.

• DJSTOCKS_60_FF_Z.dat - 60 monthly observations on the excess returns on 30 stocks from DJIA from DataStream and excess return of the CRSP index from French & Fama database for a period 01/06/2010-01/05/2015.

• DJSTOCKS_60_SP_Z.dat - 60 monthly observations on the excess returns on 30 stocks from DJIA and S&P500 index return from DataStream for a period 01/06/2010-01/05/2015.

  • DJSTOCKS_60_DJ_Z.dat - 60 monthly observations on the excess returns on 30 stocks from DJIA and DJIA return from DataStream for a period 01/06/2010-01/05/2015.
  • STOCKS_60_FF_Z.dat - 60 monthly observations on the excess returns on 20 random stocks from S&P500 from DataStream and excess return of the CRSP index from French & Fama database for a period 01/06/2010-01/05/2015.
  • STOCKS_60_SP_Z.dat - 60 monthly observations on the excess returns on 20 random stocks from S&P500 and S&P500 index return from DataStream for a period 01/06/2010-01/05/2015.

• STOCKS_60_DJ_Z.dat - 60 monthly observations on the excess returns on 20 random stocks from S&P500 and DJIA return from DataStream for a period 01/06/2010-01/05/2015.

  Description of the variables in the data sets:

• Z_1, Z_2,...,Z_20,..., Z_30 - returns of individual stocks depending on the data set.

• For calculation of the returns adjusted prices from DataStream were used (see data from Monte Carlo simulation part). Risk free return is taken from French & Fama database.

• Time period was shortened from 120 to 60 observations: 01/06/2010-01/05/2015

• Excess returns from the market and indeces:

  • Z_SP - 60 observations on excess return of the S&P500 from DataStream
  • Z_DJ - 60 observations on excess return of the DJIA from DataStream
  • Z_FF - 60 observations on excess return of the market from French & Fama database

Codes:

• load_stocks120.gss - loads the data on the returns of the randomly selected 20 socks of S&P500 and selects last 60 observations
  • load_djstocks120.gss - loads the data on the returns of the 30 socks of the Dow-Jones Industrial Average Index and selects last 60 observations
  • CAPM.prc- contains functions to estimate CAPM model by SUR and Minimum Distance methods
• CAPM.inc- sets the format for the output files from the GAUSS procedures
• CAPM_STOCKS20_FF.gss, CAPM_STOCKS20_DJ.gss, CAPM_STOCKS20_SP.gss, CAPM_DJSTOCKS30_FF.gss,CAPM_DJSTOCKS30_DJ.gss,CAPM_DJSTOCKS30_SP.gss - GAUSS procedures to estimate the CAPM models based on particular data set (20 random stocks or 30 stocks from DJIA as well as different market indexes: S&P500, DJIA, CRSP) and generate separate output files. 2019-03-05 11:51:42.893129 The replication data contain MATLAB and GAUSS codes as well as the data required for replication of the results from the paper

Our research is about robust analysis for high dimensional factor model in present of heavy-tailed data. We propose novel methods by integrating the modified Huber loss function and the common Principal Component Analysis. The methods are superior or comparable to others in numerical studies and the estimated factor number is more aligned with financial practice.

The real data in finance is from Kenneth R. French's website: http://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library.html. We use three portfolio pools: Pool A, Pool B, and Pool C to do factor analysis. Each pool contains 100 portfolios with complete monthly average value-weighted returns data from July 2016 to June 2024. The Portfolios in each pool are influenced by two primary factors. The authors have no permission to share the data or make the data public available.

We give the R codes for data generating, parameter setting and computational details in simulations.

Using all stocks listed in the London Stock Exchange for the period from January 1989 to December 2018, the dataset comprises the following series: 1. Annual returns for 20 asset growth portfolios, following Fama and French (1993) methodology. 2. Annual returns for 25 portfolios size-book to market equity, following Fama and French (1993) methodology. 3. Annual returns for 62 industry portfolios, using two-digit SIC codes. 4. Fama and French (1993) factors for their three-factor model (RM, SMB and HML). 5. Fama and French (2015) factors for their five-factor model (RM, SMB, HML, RMW, and CMA). 6. Variation of the Amihid illiquidy measure for the London Stock Exchange, following Amihud (2002) methodology. 7. Three-month interest rate of the Treasury Bill for the United Kingdom, as provided by the OECD database. We have produced these series using the following data from Thomson Reuters Datastream: (i) total return index (RI series), (ii) market value (MV series), (iii) market-to-book equity (PTBV series), (iv) total assets (WC02999 series), (v) return on equity (WC08301 series), (vi) tax rate (WC08346 series), (vii) primary SIC codes, (viii) turnover by volume (VO series), and (ix) the market price (P series). Following Griffin et al. (2010), we use the generic rules provided by the authors for excluding non-common equity securities from Datastream data. REFERENCES: Amihud, Y. (2002). Illiquidity and stock returns: Cross-section and time-series effects. Journal of Financial Markets, 5, 31–56. Fama, E. F. and French, K. R. (1993). Common risk factors in the returns on stocks and bonds. Journal of Financial Economics, 33, 3–56. Fama, E. F. and French, K. R. (2015). A five-factor asset pricing model. Journal of Financial Economics, 116, 1–22. Griffin, J. M., Kelly, P., and Nardari, F. (2010). Do market efficiency measures yield correct inferences? A comparison of developed and emerging markets. Review of Financial Studies, 23, 3225–3277.

We compile raw data from the Datastream database for all stocks traded on the Tokyo Stock Exchance, Osaka Exchange, Fukuoka Stock Exchange, Nagoya Stock Exchange and Sapporo Securities Exchange. Particularly, we collect the following data series, on a monthly basis: (i) total return index (RI series), (ii) market value (MV series), (iii) market-to-book equity (PTBV series), and (iv) primary SIC codes. Following Griffing et al. (2010), we exclude non-common equity securities from Datastream data. Additionally, we remove all companies with less than 12 observations in RI series for the period under analysis. Hence, our sample comprises 5,627 stocks, considering all companies that started trading or were delisted in the period under analysis. We use the three-month Treasury Bill rate for Japan, as provided by the OECD database, as a proxy for the risk-free rate. Accordingly, the dataset comprises the following series:

1. Japan_25_Portfolios_MV_PTBV_M: Monthly returns for 25 size-book-to-market equity portfolios, following the Fama and French (1993) methodology. (Raw data source: Datastream database)
2. Japan_20_Portfolios_MOM_M: Monthly returns for 20 momentum portfolios rebalanced in June of each year. (Raw data source: Datastream database)
3. Japan_61_Portfolios_SECTOR_M: Monthly returns for 61 industry portfolios. (Raw data source: Datastream database)
4. Japan_RF_M: Three-month Treasury Bill rate for Japan. (Raw data source: OECD)
5. Japan_C_Q: Private final consumption expenditure, in national currency and constant prices, non-seasonally adjusted, for Japan. (Raw data source: OECD)
6. Japan_Trade_Y: Trade openness for Japan, as measured by the variation rate of exports plus imports. (Raw data source: OECD)
7. Japan_RD_Y: Variation rate of R&D investment for Japan. (Raw data source: OECD)
8. Japan_IK_Y: Investment-capital ratio for Japan., determined using the methodology suggested by Cochrane (1991) (Raw data source: OECD)
9. Japan_CCI_M: Consumer confidence index for Japan. (Raw data source: OECD)

REFERENCES:

Cochrane, J.H. (1991), Production-based asset pricing and the link between stock returns and economic fluctuations. The Journal of Finance, 46, 209-237. Fama, E. F. and French, K. R. (1993). Common risk factors in the returns on stocks and bonds. Journal of Financial Economics, 33, 3–56. Griffin, J. M., Kelly, P., and Nardari, F. (2010). Do market efficiency measures yield correct inferences? A comparison of developed and emerging markets. Review of Financial Studies, 23, 3225–3277.

Sample size after data cleansing and multiple imputation procedures.

Using all stocks listed on the Japanese equity market and macroeconomic data for Japan, the dataset comprises the following series:

1. Japan_25_Portfolios_MV_PTBV: Monthly returns for 25 size-book-to-market equity portfolios, following the Fama and French (1993) methodology. (Raw data source: Datastream database)
2. Japan_25_Portfolios_MV_PE: Monthly returns for 25 size-PE portfolios, following the Fama and French (1993) methodology. (Raw data source: Datastream database)
3. Japan_50_Portfolios_SECTOR: Monthly returns for 50 industry portfolios. (Raw data source: Datastream database)
4. Japan_3 Factors: Fama and French three-factors (RM, SMB and HML), following the Fama and French (1993) methodology. (Raw data source: Datastream database)
5. Japan_5 Factors: Fama and French five-factors (RM, SMB, HML, RMW, and CMA), following the Fama and French (2015) methodology. (Raw data source: Datastream database)
6. Japan_NUCLEAR_Y: Instrument in years with a value of 1 when a nuclear disaster has occurred somewhere in the world and 0 otherwise. (Raw data source: Bloomberg and BBC News)
7. Japan_NUCLEAR_M: Instrument in months with a value of 1 when a nuclear disaster has occurred somewhere in the world and 0 otherwise. (Raw data source: Bloomberg and BBC News)
8. Japan_RF_M: Three-month interest rate of the Treasury Bill for Japan. (Raw data source: OECD)
9. Company data: Names and general data of the companies that constitute the sample. (Raw data source: Datastream database)
10. Number of stocks in portfolios: Number of stocks included each year in Japan_25_Portfolios_MV_PTBV, Japan_25_Portfolios_MV_PE and Japan_50_Portfolios_SECTOR. (Raw data source: Datastream database)

We have produced all return series using the following data from Datastream: (i) total return index (RI series), (ii) market value (MV series), (iii) market-to-book equity (PTBV series), (iv) total assets (WC02999 series), (v) return on equity (WC08301 series), (vi) price-to-earnings ratio (PE series), and (vii) industry (SECTOR series). We have used the generic rules suggested by Griffin, Kelly, & Nardari (2010) for excluding non-common equity securities from Datastream data. We also exclude stocks with less than twelve observations. Accordingly, our sample comprises a total number of 5,212 stocks.

REFERENCES:

Fama, E. F. and French, K. R. (1993). Common risk factors in the returns on stocks and bonds. Journal of Financial Economics, 33, 3–56. Fama, E. F. and French, K. R. (2015). A five-factor asset pricing model. Journal of Financial Economics, 116, 1–22. Griffin, J. M., Kelly, P., and Nardari, F. (2010). Do market efficiency measures yield correct inferences? A comparison of developed and emerging markets. Review of Financial Studies, 23, 3225–3277.

Using all stocks listed in the Australian Securities Exchange and macroeconomic data for Australia, the dataset comprises the following series:

1. Monthly returns for 20 size-price to cash flow portfolios, following the Fama and French (1993) methodology. (Raw data source: Datastream database)
2. Monthly returns for 25 size-book to market equity portfolios, following the Fama and French (1993) methodology. (Raw data source: Datastream database)
3. Monthly returns for 41 industry portfolios. (Raw data source: Datastream database)
4. Private final consumption expenditure, in national currency and constant prices, non-seasonally adjusted, for Australia. (Raw data source: OECD)
5. Fama and French (1993) factors (RM, SMB and HML), following the Fama and French (1993) methodology. (Raw data source: Datastream database)
6. Fama and French (2015) factors (RM, SMB, HML, RMW, and CMA), following the Fama and French (2015) methodology. (Raw data source: Datastream database)
7. Three-month interest rate of the Treasury Bill for Australia. (Raw data source: OECD)

We have produced all return series using the following data from Datastream: (i) total return index (RI series), (ii) market value (MV series), (iii) market-to-book equity (PTBV series), (iv) price-to-cash flow ratio (PC series), (v) primary SIC codes, and (vi) tax rate (WC08346 series). We use the rules suggested by Griffin, Kelly, & Nardari (2010) for excluding non-common equity securities from Datastream data.

REFERENCES:

Fama, E. F. and French, K. R. (1993). Common risk factors in the returns on stocks and bonds. Journal of Financial Economics, 33, 3–56. Fama, E. F. and French, K. R. (2015). A five-factor asset pricing model. Journal of Financial Economics, 116, 1–22. Griffin, J. M., Kelly, P., and Nardari, F. (2010). Do market efficiency measures yield correct inferences? A comparison of developed and emerging markets. Review of Financial Studies, 23, 3225–3277.

Data contains information from ESG Corporate Ranking by RAEX from December 2020 to February 2022, its dynamics, RSPP indices "Vector of sustainable development" and "Responsibility and openness", data for the Fama-French model, market capitalization and assets of companies and daily stock closing prices.