Fundamental beta and portfolio performance: evidence from an emerging market

ABSTRACT

The market beta is decomposed into fundamental and bubble beta to assess their effectiveness in the portfolio performance in both static and dynamic time-varying frameworks. The empirical results from India on 12 sectoral indices with NIFTY 500 as the market index establish that the portfolio constructed using the fundamental beta proportions performs better than the naïve, Markowitz mean-variance, market, and bubble beta portfolios with larger Sharpe ratio in both the static and dynamic time-varying estimates. These results open up far-reaching implications for investment analysis and contribute to the recent literature that combines fundamental analysis in the construction of portfolios.

KEYWORDS:

• Portfolio optimization
• sharpe’s single index model
• bubble CAPM
• time-varying Beta

JEL CLASSIFICATION:

• G10
• G11
• C61

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. Important limitations highlighted in the literature include, mean-variance preferences fails to be monotone and using variance as risk measure have disadvantage as it treats both profit and loss equally, while the risk is associated only with the loss. Refer to Maccheroni et al. (2009) and Ahmadi-Javid and Fallah-Taft (2019).

2. Subsequently, the Capital Asset Pricing Model (CAPM) developed by Sharpe (1964), Lintner (1965) and Mossin (1966) became the benchmark in evaluating the price and risk premium of any security.

3. In another stream of research, fundamental beta is derived as the estimated value from a cross-sectional regression of historical betas of stocks with fundamental ratios (see Rosenberg and Marathe 1975).

4. This model is preferred over the Kalman filter, for a similar exposition of the bubble CAPM of Anderson and Brooks (2014) in a time-varying framework.

5. This is the range of risk-free rate in India during the sample period and the same risk-free rates are used for equal weight and Markowitz mean-variance portfolios as well.

6. The results remain similar for the risk-free rate at 4 and 8 percent as well, which is not reported here but can be obtained upon request.

7. It shows only first-order autocorrelation and subsequent orders show no autocorrelation.

8. We used ‘sharpeTesting’ package in R for the estimation.

Additional information

Notes on contributors

Hafsal K

Hafsal K is a Research Scholar in the School of Economics at the University of Hyderabad, India.

S. Raja Sethu Durai

S. Raja Sethu Durai is a Professor in the School of Economics at the University of Hyderabad, India.