Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Strategies & Market Trends : Befriend the Trend Trading -- Ignore unavailable to you. Want to Upgrade?

To: TradinSOB who wrote (16042)	2/20/2001 6:21:27 PM
From: Ken Adams	Read Replies (1) \| Respond to of 39683

So, this reflects all trades... both long and short? IOW, the bottom of the "profit" list shows some really fab results that obviously fit the 5e/8s profile? What was the criteria for the original list you fed your backtest? Nice work, BTW! Ken

To: TradinSOB who wrote (16042)	2/20/2001 6:26:56 PM
From: smchan	Read Replies (1) \| Respond to of 39683

Anything else? 1) Kill the commissions. We all trade with different brokerages and taking out commissions skews the profit numbers. I'd rather account for commissions in my head. Remember, the idea behind backtesting is to find good future trading candidates; not gauge how a system would've EXACTLY performed. 2) Add number of winners/losers and percentage of winners/losers. Max and average profit and loss per trade is also useful. 3) Shorten the backtest. General feeling here is anything over a month is not useful. 4) Check your logic. It's difficult for me to believe MAX would've performed so well on so many stocks. In my longer term backtests, there are only a handful of decent winners. It's highly suspect that your backtest has identified so many winners with such high profit. I suggest a random sample against a hand backtest with charts. 5) I usually only test a fixed number of shares but I like the equity approach too. 6) Thank you for sharing you work. Please continue to do so! :-) Sam

To: TradinSOB who wrote (16042)	2/20/2001 9:05:40 PM
From: Steve_GTS	Read Replies (2) \| Respond to of 39683

You may want to check your calculations. I ran 2 independent programs and they confirmed each other, where your program results were different. Here are my results for your best stock CLS Steve 5e8s System Inputs cls Symbol 60 Bar Periods (1, 2, 5, 10, D, etc.) 510 Bars Tested $20 Commissions/Slippage Round Trip 200 Shares Traded 5 EMA Bar Period - Fast Line 8 SMA Bar Period - Slow Line 11/02/00 14:30 Start Date 02/20/01 15:30 End Date 78.5 Days Backtested 510 Bars Backtested $14,186.00 Initial Investment $24,826.00 Current Account Value $25,926.00 Account Maximum $14,008.00 Account Minimum ($178.00) Max. Drawdown ($1,120.00) Total Commissions/Slippage R.T. $9,648.00 Closed Trades Profit $8,528.00 Closed Trades Profit with C/S $56.05 Last Close Short Current Position $108.69 Profit Per Day with C/S 0.7 Trades Per Day Total Closed Trades 56 Total Trades $9,648.00 Total Profit $172.29 Avg. Trade Profit $19,766.00 Total Gains $732.07 Avg. Gain $3,150.00 Max. Gain ($10,118.00) Total Losses ($348.90) Avg. Loss ($882.00) Max Loss 2.1 Avg Gain/ Avg. Loss 3.6 Max. Gain/ Max. Loss 27 Winning Trades 29 Losing Trades 48% Win/Total Trade % 59 Longest Gain, bars 17 Longest Loss, bars 5 Consecutive Gains, Trades $4,576.00 Consecutive Gains, $ 4 Consecutive Losses, Trades ($2,278.00) Consecutive Losses, $ 1.25 Cons. Gains/Losses Trades 68.0% Period Tested 221.0% Annualized Long Closed Trades 28 Long Trades $3,832.00 Total Long Profit $136.86 Avg. Trade Profit $8,972.00 Long Gains $640.86 Avg. Gain $1,426.00 Max. Gain ($5,140.00) Long Losses ($367.14) Avg. Loss ($868.00) Max Loss 1.7 Avg Gain/ Avg. Loss 1.6 Max. Gain/ Max. Loss 14 Winning Long Trades 14 Losing Long Trades 50% Win/Total Trade % 23 Longest Gain, bars 10 Longest Loss, bars 27.0% Period Tested 87.8% Annualized -21.0% Buy and Hold Period -68.2% Annualized Short Closed Trades 28 Short Trades $5,816.00 Total Short Profit $207.71 Avg. Trade Profit $10,794.00 Short Gains $830.31 Avg. Gain $3,150.00 Max. Gain ($4,978.00) Short Losses ($331.87) Avg. Loss ($882.00) Max Loss 2.5 Avg Gain/ Avg. Loss 3.6 Max. Gain/ Max. Loss 13 Winning Short Trades 15 Losing Short Trades 46% Win/Total Trade % 32 Longest Gain, bars 6 Longest Loss, bars 41.0% Period Tested 133.2% Annualized 21.0% Short and Hold Period 68.2% Annualized

To: TradinSOB who wrote (16042)	2/20/2001 9:21:35 PM
From: Dr. Stoxx	Respond to of 39683

Great work, SOB! Thanks...TC

To: TradinSOB who wrote (16042)	2/21/2001 9:27:10 AM
From: TradinSOB	Respond to of 39683

BTTT-MAX Backtest I did find one error: a tiny, tiny difference between the calculations of present averages and past averages, which rarely caused a crossover to be missed (when the value of the difference was less than about 1/100 of a cent). Fixed that and got new results, posted: home.earthlink.net home.earthlink.net These results still don't agree with Steve's for his example CLS. Two of my intermediate files are available for CLS: home.earthlink.net is the cross-over detect output and home.earthlink.net is the profit calculation per crossover (which is done differently for the first and last crossovers).