Is Socially Responsible Indices Weak Form of Efficient Market? Evidences from Developing Economies

The study is an attempt to identify the presence of randomness in the socially responsible indices (SRI) of the stock markets of developing countries. Five developing economies are considered for the test of randomness on daily, weekly, monthly, quarterly and semiannual return of socially responsible indices and their benchmark indices. Shapiro-Wilk test is used to test the normality of the data whereas Runs test and Augmented Dickey-Fuller test are used depending on the randomness of the data. It is observed that India, Arab and Egypt show non-randomness whereas Brazil and South Africa show randomness in daily returns. Weekly returns on the other hand are random in Brazil, Arab, South Africa, and non-random in India and Egypt. Monthly and quarterly returns show randomness in India, Arab, Egypt, South Africa and non-randomness in Brazil whereas semiannual returns show randomness for all economies. It is also observed that most socially responsible indices resonate the randomness patterns of their benchmark indices. Most of the non-randomness is seen in short-run indicating inefficiency in the market. However, in long-run, the market goes random or efficient which is an indication that more than average profit can be earned by resorting to technical trading in the short run. Moreover, the similarity in randomness between socially responsible indices and their benchmark indices indicates that similar trading strategy can be applied by traders in both these indices to garner profit.

Exploration into power of homogeneity and serial correlation tests

Verification of regression models is primarily based on analysis of error terms and constitutes one of the most important steps in applied regression analysis. In cross-sectional models, the error terms are typically heteroskedastic, while in time series regressions the errors are often affected by serial correlation. Consequently, in this paper, we focus on Monte Carlo simulations applied to explore the power of several tests of homogeneity and tests for presence of autocorrelation. In the past decades, the computational power has increased significantly to allow the benefit of simulation from exact distributions, which are not defined explicitly. We will discuss 1) testing of homogeneity for a given number of components in the exponential mixture approximated by subpopulations and 2) simulation of power in several commonly used tests of autocorrelation. For the first case, we consider exact likelihood ratio test (ELR) and exact likelihood ratio test against the alternative with two-component subpopulation (ELR2). In the second case, we consider the Durbin-Watson, Durbin h, Breusch-Godfrey, Box-Pierce and Ljung-Box tests of 1st order serial correlation and the runs test of randomness in two different types of linear regression models.

A Test of Randomness for Finite Spatial Distributions

A statistical method is presented for determining randomness of points spatially distributed in two-dimensional space. The procedure is based on a distance-to-particle (nearest neighbor) model derived from an elementary Poisson process. In a previous derivation of the method, an extension to the model was proposed and used without adequate empirical justification. Herein the test is derived in detail and its performance evaluated with Monte Carlo simulations. Results indicate that the model extension provides adequate representations when the null hypothesis is true.