You are apparently confused about what an MA is. Because it reflects the average price (close, open, whatever) of an instrument over time, it is a lagging indicator. It trails current price. It only indicates what has happened. Past tense. Under what circumstance would you want to trade an instrument that reflects an MA? Makes no sense at all. And stock777 is correct. The only way to replicate an MA is to buy the close each day, assuming that your MA is based on close.
As pointed out earlier, that doesn't replicate an MA. If you're trying to replicate a 5 day MA - after you've bought the close for the first 5 days, what are you going to do on the sixth day? If you buy the close on the 6th day, you now have an average cost that matches the 6 day MA, not the 5 day MA. Constantly buying the close just gives an average cost basis over N days. But unless you can sell the shares you bought N days ago at the price you paid for them while you buy the new shares at the current close - you can't maintain an N day MOVING average. Looking at a short MA on a chart doesn't say anything. Still have no idea dale's thinking the benefits of buying/selling at the MA would do for him. Is he thinking he can predict the MA better than the price? If he's thinking he could arb his "MA security" against the physical security, that seems very doubtful for many practical reasons. Since he started the thread asking about it, it would be nice if he'd stop with the "just look at a chart" stuff and tell people why he thinks it would be so great if such a magical instrucment could be created.
you do what the MA calculation does: you get rid of the positions outside the MA window you're interested in. in your specific case, at some time during day 6 you sell what you bought on day 1 and buy the close of day 6. your cost basis will then exactly replicate the MA.
I also don't get the idea. Maybe you're not communicating the idea sufficiently because you can't or won't, but the closest this sounds like to me is Marty Schwartz's idea of 'buy long when the price is over the EMA(10) and exit when it's under the EMA(10)'. Otherwise, realize that using very short MAs can give you an incorrect perspective of the major trend, and as others have pointed out, MAs print after the fact.
The MA calculation just drops off the sixth day as if you dumped the shares at the price you paid for them - so there are two ways of looking at it - a moving position or a bunch of separate positions: Latest 6 days prices - 11, 10, 9, 8, 7, 6 MA of last 5 days is (10+9+8+7+6)/5 = 8 Your buy @ the close: Buy 100 @ 11 = 1100 (day 1) Buy 100 @ 10 = 1000 (day 2) Buy 100 @ 9 = 900 (day 3) Buy 100 @ 8 = 800 (day 4) Buy 100 @ 7 = 700 (day 5) Sell 100 @ 6.5 = -650 (assumes you sell at better than the 6th day close) Buy 100 @ 6 = 600 (day 6) Average cost basis on the overall 500 share "moving position" is 8.90/share Alternatively, you don't treat it as a "moving position" but rather as 5 separate positions and take daily P&L hits for dumping the oldest position and replacing it with a new position - presuming that's how you were thinking about it. So in this case you'd record a P&L hit of -$450 and then assume your cost basis on the remaining 500 shares (average of 5 individual lots) is the 5 day MA of $8/share. In which case in a downtrending market you record a stream of daily losses to keep the net moving position artificially tracking at the 5 day MA. But frankly I don't see why the overall approach in general (regardless of whether you roll the cost averaging for the moving position or you post the net rollover P&L each day) is of any special value. And what's the point of starting the thread if the originator is going to play games about explaining why/what he's thinking of?? People shouldn't bother asking such a question if they're unwilling to explain it sufficiently for people to understand the context. "Just look at a chart with a short MA and figure it out"?? That's an assinine response when he's the one asking for help in the first place. Sounding a lot like a junior trader wannabe who thinks he's found the holy grail, wants others to give him answers, but doesn't want to explain the context of his questions for fear his "secret" will get out. Yawn.
He asked a very simple question, and you are all over-analyzing it. Not that I know the answer either.
that's a little uncalled for, don't you think? perhaps he's playing devil's advocate and thinking through the implications of *not* being able to constructe an MA-equivalent synthetic. regardless, is the tone and language of your post really necessary?
After thinking about it for a sec i realized that it would be impossible to create an instrument that traded like the MA, why? because if you could, you could buy a the 200 MA of a stock that was in a 6month uptrend, the MA of the next day is going to be higher than the MA today, unless the stock drops like 95%, because you are replacing a value 200 days ago with the current value, which will almost assuredly be above the price 6 months ago, for a stock that has risen. That would make the investment a winner 99.99% of the time, and such things do not exist, maybe im wrong and if someone figures it out they would be quite wealthy. Anyway, I was looking at my daily charts which i use a 13 day ema on and (btw, i do follow schwartz's envelope) and made a simple observation that price of the stock and its MA always do eventually meet, and its ALWAYS, so its a sure winner if you arbed the two, like i believe someone mentioned on this thread earlier. thats it and thats all folks
well, i was being selfish, if there was a way, it would have been the greatest thing pretty much i've ever heard of, and i didn't want to just give it away. that said, some of you need to calm down.