Currently testing if monthly fund characteristics (size, capital flows, age, risk, persistence,...) explain funds abnormal returns.
My data is set as a panel with 1000 equity mutual funds over the period (2000-2017).
Therefore, I run the following regression
Alpha = intercept + B1(fund size) + B2(capital flows) + ... + u
Most past literature such as Chen et al. (2004) and Carhart (1997) use the Fama MacBeth procedure to test such relationship.
However since my dataset suffers from both time series and cross sectional correlation. I am better off (according to Petersen (2009)) by using a fixed effect regression and cluster residuals by fund and time to adjust standard errors.
Anyway, I run the regression using both models (fixed effect and Fama MacBeth procedure) and I get slightly different results.
I was just wondering what would be better model to tackle such problem.
References: Chen, J., Hong, H., Huang, M. and Kubik, J.D., 2004. Does fund size erode mutual fund performance? The role of liquidity and organization. The American Economic Review, 94(5), pp.1276-1302.
Petersen, M.A., 2009. Estimating standard errors in finance panel data sets: Comparing approaches. The Review of Financial Studies, 22(1), pp.435-480.
No comments:
Post a Comment