Of course it's alot of work. Trading is alot of work. Give us ideas of how to read price action accurately if you have a better idea. Because nobody really understands what price action is or how to accurately read it. It's impossible. But if you know how, then let us know.
=============== FEB28-07; I dont mind work, but as you hint thank God trading isnt that much work; like 50dma [written down work]as a guideline. Meaning above 50dma is buy; below 50dma is a sell usually.Remember a liquid NYSE oil stock that kept trending weak,down when most were up.So had enough descretion to short it well before it ever got to 50dma. Like moving averages better than straight trendlines; they require less discretion.
====================== DaTrader; I didnt understand much of what DR.Koch wrote also ; and frankly that doesnt bother me a bit Agree''every day you moving averages will change''[except 50 dma doesnt change much daily ] But i never adjust a 22 ema period moving average. ---------------------------------------------------------------------------- I am not sure what you call a false reading, mayde a sideways trend[narrow range-flat moving average]; usually dont trade a sideways trend [knowingly anyway]-laugh out loud. Trading can be difficult, but 12 hours a day/ work 6 days a week helps, & i consider this work also . But dont need to adjust a 22 period EMA .It auto adjusts constantly enough,typical EMA.[except sideways trends] Also wouldnt limit entry/exit to 50dma ........
We use butterworth filters in speaker design circuitry (HPF, LPF, BPF) helps keep lows from tweeters and highs from woofers and lows and highs from a MIDrange driver. Or EQ circuits. Now how that applies to deciding whether it's a good time to enter a position or exit one.... Hmmm....
You could always use your ma's to gauge divergent bar patterns from a mean. You could look at price action around ma's to help gauge "sentiment" (probable direction) of a bar pattern. You could use a band of ma's as a replacement for fib levels, or to tell you when price is rangebound. And if none of those things help understand price action, you could always use one as a crossover entry, to help get in if price reading fails you. And if none of those things help read price action, you could always use one as a stop loss level. Sure, there just lines on a chart. But by definition, price hits them, or not-doesnt that tell you a heck of a lot about price action?
==================== That tells me a lot about price action And they can also be good finders of support & resistance levels; sometimes amazingly so, sometimes not, so i watch for both patterns.
If you already make good decisions based off EMAs or SMAs, a Gaussian filter will only improve your accuracy. All of the aforementioned MAs are low-pass filters. The only difference is the effectiveness at keeping low frequencies intact.
I use ma's for only one part of my analysis and that's to see what the herd is looking at. I like the NN software with the "nearly 80% accuracy" that one may buy for only $19700. This is for 19 Categories. WOW It's not that the modules are not robust, no,no they are specifically tailored to your specific market instrument. The cutting edge technology of neural networks magically transforms this: Two 2 moving averages 1 is a simple 10 MA line 2 is a Exponential 5 day line Into this: $$$$$$ All for only the cost of an economy passenger vehicle. Gotta love it! Now bend over and bark like a dog.
I attach trading simulation results obtained using Standard And Poors 500 index tracking stock (symbol SPY) historic daily price data. Data totals 13.97 years from 29 January 1993 to 26 January 2007. I use closing price to avoid conflicting trading signals. This simulation uses exponential moving average time constants from 10 to 500 days. Trading is long positions only. Position size equals (1 % of account equity) / (10 times the 20 day average true range). All tests show profitable outcomes. Longer term methods show greater profit.