Abstract
When analysing the data periods including the pre-financial and financial crisis periods, the results show that investors might make profits by using Variable Length Moving Average (VMA) trading rules as buying signals rather than as selling signals shown for the Brazil, Russia, India and China (BRIC) stock markets. However, investors may find it difficult to make profits in a financial crisis period, suggesting that more detailed information should be investigated, since the significant results shown during the full period might not reveal the differences between the pre-financial and financial crisis periods.
Notes
1 The 5, 20 and 60 present a week, a month and a quarter for the 5-20, 5-60 and 20-60 trading rules. In addition, the 5-20, 5-60 and 20-60 trading signals are emitted as the weekly MA rises above (falls below) the monthly MA (5-20), the weekly MA rises above (falls below) the quarterly MA (5-60) and the monthly MA rises above (falls below) the quarterly MA (20-60).
2 In the 5-20 trading rules, the practitioners have selected a 5-day moving average as the SMA and a 20-day moving average as the LMA. The 5-60 trading rules indicate that practitioners have selected a 5-day moving average as the SMA and a 60-day moving average as the LMA. The 20-60 trading rules imply that practitioners selected a 20-day moving average as the SMA and a 60-day moving average as the LMA.
3 Didier et al. (Citation2011) mention that the 2007–2008 financial crisis origins were in the US subprime housing finance market.
4 The value of the US subprime mortgages was estimated at $1.3 trillion as of March 2007, with over 7.5 million first-lien subprime mortgages outstanding. With the availability of easy credit conditions, there is evidence that both government and competitive pressures contributed to an increase in the amount of subprime lending during the years preceding the crisis.
5 The VMA rules could avoid data-snooping biases as mentioned by previous studies (Brock et al., Citation1992; Bessembinder and Chan, Citation1995; Ranter and Leal, Citation1999; Coutts and Cheung, Citation2000; Kwon and Kish, Citation2002; Tian et al., Citation2002).