Replies to post #533 on Andalay Solar Inc fka WEST

[dmlabuda]

I came up with the values I use to compute my indices after a lot of back testing and research. I could never understand why the 50 and 200 day moving averages were selected and other than popularity with the rest of the world (it's nice to know what others may be thinking), I never found them all that useful. The RSI set at 14 never seemed to trigger anything much and being a swing trader, I wanted a more useful and active indicator for my purposes. I recall reading something somewhere about Wilder choosing the number 14 because it was one half of the lunar cycle. But, that makes no sense to use 14 because the price charts are based on trading days and not calender days. So, I came up with my own base numbers for the RSI, Stochastics and %R that are based on the lunar cycle in terms of trading days and knowing that there are actually more than 12 lunar cycles in a year. While back testing the results, I wondered if I could come up with a more meaningful moving average to identify support and resistance levels and using a little conjecture thought it might be interesting to base a moving average on 1/3 of the trading year testing the "everything happens in threes" theory. So, the 84 day EMA was born and the results of testing with it were truly amazing. So, I now use it in conjunction with the 50 and 200 day MAs and also the Lunar and bi-Lunar exponential moving averages of 20.38 and 10.19 candles respectively. What I found was that 10 and 20 day moving averages (although they are not precise) do, identify support and resistance in a trending market very nicely and that the 84 day EMA is amazing at identifying support and resistance that the 50 and 200 day MAs miss. So, that is why I use the math that I do in calculating the indices and averages displayed on my charts. For a long time I could not find a better base than 12, 26 and 9 for the MACD, buy I have been experimenting with a 10,20,5 MACD that would make it somewhat Lunar and my initial research seems to prove that I am getting better signals sooner using those numbers.