jpb Veteran   Posts: 168 Joined: 5/11/2005 Location: Brown Deer, WI  User Profile | I really like the analysis spreadsheet data the more I look at it. I noted earlier that there is a Long vs Short. I didn't see in the spreadsheet how it determined that so I'm suspecting Mark is using a Macro to run through the data and dynamically build the Analysis sheet. I also see on the Trades tab that Mark overcame the text based fields for G/L($) and Entry Date that come out of Portfolio Simulator by using the =VALUE() function -- very nice. I've been conducting a change all with RegEx (inside of LibreOffice) to convert those to proper numbers/dates but I'm going to remember this trick. One observation while I was comparing Mark's analysis against my pivot table, specifically in the Default and Bear market state. Two strategies "ATM T3-S3 MOC In 224" and "ATM X ULT MOC In 224" both show a positive G/L(%), arrived at by summing all the G/L(%) for that strategy in that market state. However, if you sum the G/L($), you would see significant losses. T3-S3: G/L(%) 86.08% G/L($) -325,059.02 X-ULT: G/L(%) 16.52% G/L($) -69,198.08 I dug into the trades found on the Trades tab for T3-S3 and see that on 8/20/2015, 8/21/2015, and 1/5/2016, this strategy had a few bad days. Because we are using % Equity as the allocation, the quantity acquired caused significant losses in a short period of time. Usually I'll look just at the % since it normalizes value, but since the loss was significant (without looking at the data), I had originally thrown out those strategies even though they demonstrated good returns (%). But looking at the data, it tells me I forgot why we look at %. If I pulled out the trades from those 3 days, the picture changes dramatically: T3-S3: G/L(%) 352.16% G/L($) 621,076.68 I like that Mark removed the statistics that can cause distraction (at least distract me). Thanks, Jeff B. | |
