1. Small Sample Properties of Non-Parametric Tests of the Martingale Hypothesis.
- Author
-
KUMAR, DILIP and MAHESHWARAN, S.
- Subjects
STOCK prices ,MARTINGALES (Mathematics) ,EFFICIENT market theory ,RANDOM walks ,MONTE Carlo method ,STOCK exchanges ,FINANCIAL market reaction - Abstract
The central goal of this paper is to study the small sample properties of non-parametric multiple variance ratio tests and the runs test and to apply suitable tests on the daily data of the stock prices of Indian sectoral indices to study their martingale behavior. We have also used moving a subsample approach to examine the dynamic behavior of the test statistics. This helps us to identify the sensitivity of results to a particular sample period and to obtain inferential findings robust to possible structural changes and presence of influential outliers. Joint sign test turns out to be better than other tests for sample size less than and equal to 1000. Our results provide evidence against the weak-form efficiency of all the sectoral indices, except for CNX IT. According to our analysis based on the moving sub-sample approach, except for BANK NIFTY and CNX INFRA, all the other indices have become more efficient after the sub-prime crisis. [ABSTRACT FROM AUTHOR]
- Published
- 2014