This is an idea that I have never tried but I have noted previously. I am curious how you are determining an uptrend, ie., the direction of a moving average, higher highs and lows, etc. Thanks.
Yes. I like to play long-term moves, holding for anything between several months to a year or more (on the winners - obviously I bail from losers a lot faster!). Prefer currencies, stock indices, and interest rates, but I'm not averse to playing commodities if there's a clearly defined bull or bear market. Generally I find there are at least 4 or 5 major bull or bear trends per year - if you catch only 3 of them and make 10% on each, then that's a pretty good return on cash. Add in some leverage or good entry & exits, and you can do pretty well. The downside is that it only takes a couple of big mistakes to screw up your year, or miss out on what would have been a huge winner.
YES! If I'd liquidated on the close yesterday I would have missed the premarket EXPLOSION today. That's why I held it!!!!!! "FORESIGHT = HINDSITE in reverse"! -..... T-REX
...I owe you some money. How do I wire your bank account? Saying you have tested hundreds of such strategies is absolutely priceless information. I have tested maybe 20 so far with the same results. I keep grinding them out in the hope that there may be something obscure there. Even the ones which are marginally profitable over long periods exhibit horrible equity curves. Re the buy open sell close strategy, it worked through the recent pull back. I don't trade it, of course, more to be made in shorter timeframes.
Here is a very simple swing system that should demonstrate to anyone that there is value in holding overnight, and value in buying pullbacks in a trend. The following system was run over the past 12 months, Oct 9, 2002 to Oct 9, 2003. BUY RULE: Create a list of the DOW 30 stocks. Log the daily stochastic (RawK, 8 period) of each stock. When the overall AVERAGE RawK of all 30 stocks is below 30, buy all the next morning. (This is a good way of determining a market bottom, for any trading system that goes long.) EXIT RULE #1: Do nothing for one day. The first day that the stock closes up, AND closes above your entry price, sell at the close of that day. or EXIT RULE #2: If stock has not sold after 5 days, wait until stock reaches entry price, then sell at the close of that day. RESULTS: 337 trades. 79% profitable. 2.16 Win/Loss ratio. 225 net points gained. Average hold for winners: 4 days. Note that this system did not specify the conditions on each stock at the time of purchase. It bought only on the average of the Dow 30 stocks. You can use your imagination, or your backtesting software, to comprehend the gains if you had looked at a larger universe of stocks for better opportunities when these same conditions existed. You may also note, if you chart this system, that these conditions occured regularly each month. And each time this occured, it was an excellent buying opportunity. And who said that technical studies are useless in a random market?
Hi lindq. I have a ? about that system. Exit rules say if stock not sold after 5 days mean if stock has not closed up since you bouht it correct? If that is the case what if the stock stays down for a while. An example would be something like TYc or EK which has stayed down for extended periods. Is that part of the losses in the stats and marked to mkt since they have not been liquidated? Thanks
Please use your own exit rules as you see fit. I only posted this as an example of what can be done with a simple swing system. I don't trade this system, but I do track it as an indication of when to heavy-up on long positions. You may find it of value on its own, or as a backdrop to another long swing system.
Thanks for all the responses - good to know that I am not the only one of my kind! Another question for my EoD brethren: how do you assess system performance when testing? My preference is for things like Ryan Jones Expectancy ($Return for $Risked per trade) and the annualized version of the same ($Return for $Risked per annum). Any (polite) suggestions? TIA!