I am new to Portfolio analysis. I have 1 system with 8 assets and 2nd system with 5 assets. What can I do to help analyze the results? Individually, both these systems do well. Just looking for insights into what I can do to help with analysis. For position size, I use percent of equity of 100 / # of assets traded in the system, with a little less to eliminate NFS. Is there a way to combine analysis to show an equity curve combined for both systems? Along with a combined metrics report (probably has to look different because of 2 different systems and different assets traded. Maybe there is an excel to help with this? Your thoughts would be appreciated. Thanks, Larry
Hello Larry, Why not keep things simple, and bet on 1 or 2 trading systems and scale up size for the rest of your life? I think a portfolio of systems is too much work for little reward for the retail at home alone trader/investor. Then you have to spend all your life managing about +20 systems Find 1 or 2 trading systems you believe in stick with is what I am doing to scale.
Check out https://strategyquant.com/quantanalyzer/ You can segregate strategies and get stats by specific markets
Alternately, you could use Multicharts Portfolio Manager feature and send the trade stats to a CSV file: Portfolio_NetProfit Portfolio_NumLossTrades Portfolio_NumWinTrades Portfolio_TotalTrades Vars: TotalTrades(0) ; TotalTrades = Portfolio_TotalTrades; If TotalTrades > TotalTrades[1] then // new trade Print(file("C:\port\portfolio.csv"),"Trade num:",TotalTrades," Net=",Portfolio_NetProfit);
simplelikeme wrote: “Find 1 or 2 trading systems you believe in stick with is what I am doing to scale.” [snip] This what I am planning to do. 1 system is based on US assets and I am looking at Country ETFs/ other assets that are not correlated for diversification. My ask is just to have 2 backtested systems combine their equity curve with related metrics. I am trying to see what more or less risk I obtain running 2 systems with non correlated assets. Thank you, Larry
You are likely to see less risk, but also a lower return. I believe most of the risk reduction comes from the number of instruments you are trading. The other thing you need to do is optimize the betsizing for each instrument.
Syswizard, How would you optimize betsize for each instrument. What would you be looking at. That seems to me to make the system less robust. thank you, Larry
Analyze the equity curve for each instrument/strategy combination. Use either the slope of the equity curve or use a cross-over of the equity value vs. the average of the equity value. Why would you say that ? Betsizing can make a mediocre system look like a champ.
Here is what I thought, When picking a possible asset for a portfolio, I test system just against that asset and just choose 90% percent of equity to trade to look at assets to trade. Then I trade betsize as 100 divided by # of assets minus slippage. Are you testing betsize as a variable? For each asset you are looking at slope of equity curve profit to help determine betsize? Thanks, Larry
Treat the systems as assets. So you have a portfolio of 2 assets. You can run minimum variance, max returns,... given some constraints