Content area
Abstract
This thesis tests the validity of the weak-form EMH on the Canadian TSX equity market. Qualitatively, the efficiency on the TSX is compared against that of the NYSE, the NASDAQ, the LSE, and the TSE. Quantitatively, a variety of statistical tests are used to test for the randomness of return series. Results of the autocorrelation, the BG, the runs tests all suggest that returns are serially correlated, except returns on the TSX 60 capped index. Univariate unit root tests (i.e., ADF, PP, KPSS), reject the RWM. Non-linear dynamics tests are also applied. BDS results reject an IID underlying residual series indicating that a noisy non-linear process describes the data. Finally, results of the R/S analysis, indicates that all TSX index returns possess long-memory properties of an anti-persistent trend-reversing behaviour. Overall, results uniformly reject the RWM governing TSX equity index returns, implying that the Canadian equity market is weak-form inefficient.
Key Words: Weak-form efficiency, weak-form efficiency tests, RWM, equity returns