Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
Myst, It would be nice if someone would do a "walk forward" and post it here over a few days to generate questions and instructions from the master. Any volunteers out there?
whitelake
Myst,Jerry, Myst said So if you do a walk forward test for the way I intended the program to work, will it be a valid (and accurate) assessment of its functionality and performance?
This is not my reason to want to do a walk forward. I see the validity of Xdev as a tool, and just want a way to simulate running an actual trade until Xdev 3.0 comes out. It's the re-opts that confuse me most. How do you reconcile the recommended trades for the future with the actual past trades. With this market Xdev 3.0 will be long out before I see any cash to try it with anyway.
whitelake
Ervin, I tried starting with a stock that I new nothing about so as not to prejudice the walk forward. The walk forward was to start in march of 2001 so I optimized back into 2000. But when I started the walk forward I entered "the land of confusion". Can you suggest a method? Those re-opts have me confused as to how to make them physically work with XDev, and I have never seen it posted exactly how to do it either.
whitelake
Mark, That BRCD looks like what I hope to see in my future. Thanks for sharing them. Do you have any actual buys with XDev yet?
whitelake
PCA comparison with JBL-----I tried to use as close to the same settings as possible in Myst's 2nd example to see what AIM might do with JBL.
It's not even close with 59.6% for XDev and .3% for AIM. The small difference in B&H is the way the actual start dates are figured in each program. A more experienced AIMer could surely beat my set up in PCA.
whitelake
Myst, Optimization period question? If you use 5 months initially would you always use 5 months through the life of the play at each re-optimization. I was wondering if a major trend reversal would cause you to re-opt with a shorter time span?
whitelake
Bean, I was able to get only a 1% gain with 5.5,1.5,50,50,5,-3
Resulting in 50.46% gain.
The big difference is you would be 65.95% in stock and I would be only 39.35%. Which would be better?
whitelake
Myst, It is set up for >12 and not 12 or more.
whitelake
Keith, Did your get your problem solved with ADBE?
http://www.investorshub.com/boards/read_msg.asp?message_id=391748
whitelake
Myst, I probably should have added this to my last post.
I was inspired to ask the question because of the problem Lemonhead was having starting a trade with ADBE. You may have answered the question and I missed it.
whitelake
Myst, Is there a problem with "you must have a time period greater than 12"? Should that be "greater than 11"?
whitelake
Tom, Parish in the Bayou State is the same as county in your neck of the woods.
whitelake
Playtime, I'm sure by now you saw results on HD from MYST that were similar to yours except for one thing. You had a lot of really small trades which is good for the brokers. It seems there is some skill to be acquired here.
whitelake
Myst, I see there is some skill to the optimization. HD was picked out of the blue for practice. The first thing I noticed, beside your 9% advantage, was that I am 93% stock and you are 8.9% stock. Would you think that things like that would be considered, or just go for the max settings and don't look back?
whitelake
Myst, Dialup here with no sign of a lost connection. It might just be a connection problem though, and I guess we'll have to give it some time and keep and eye on it.
I got 2,5,22,22,-1,-3 for a loss of(8.82)% vs (20.66)% B&H when I ran HD if anybody is into comparing notes. That was start 1/18/02 and end 6/18/02 for $10000.
whitelake
Myst, I was about 5min into optimizing HD and when I made a change the hourglass stayed on for about15 sec the XDev did it's own exit. Poof it was gone. It's done this before, and I'm wondering if anyone else has this happen. Could be telling me that HD is a loser.
whitelake
Keith, If you get bored and want to compare optimized results try HD, $10,000 and start 1/18/02. I am trying it till the cool sets in so I can load the ole boat.
whitelake
Dick, I just opened the .exe file the same as the original XDev2.0 file that we downloaded, and it is set up to put it in the same directory. Then I throw the old file away, and I don't even know if that is needed. Then I tossed the old desktop shortcut in the trash and created a new one to point to the new file. It was easy or I'm just lucky this time.
whitelake
Myst, 2) just download the executable file and replace it in the same directory as the old executable file This worked fine for me. It's nice to see the number of shares for the first buy.
Thanks
whitelake
Matt, Vermilion Parish here----How about you?
whitelake
Keith, Your way would work, but I would need something faster and simpler which is what I believe Myst will have for us tomorrow. I see what you mean about the optimizations. I have noticed that 2 or 3 people can get a little different set ups, but remember that re-optimization will have to occur along the way anyway.
whitelake
Myst, You said thanks for your patience, it is very much appreciated. It's not hard to be patient with someone as fair and helpful as you. The free upgrade is going the "extra mile".
whitelake
Keith, What are your thoughts on optimize and save so many times? I am missing something here maybe, as it would seem to me that after you had done your best, doing it again wouldn't help. Did you give a look to my idea of optimizing back 5 or 6 months then save. Then readjust the start date forward so that 12 days of data will show on the chart, and then going from there live.
Myst has said a couple of posts back that he has some ways to do this while we wait for 3.0 which he plans to speed along. I am anxious to see what he proposes.
whitelake
Dick, I would optimise QQQ back about 5 months or so then save. Then I would put start date on May 31,2002. Now you have your next buy price. If it starts up and makes a long mad run it will never give a buy till it starts down and hits the lower band again. As far as buys you missed I guess they are missed. Try that and see what you think.
whitelake
Richard, I think you have to be showing at least 12 days so the 12 day sma can compute. Try a start of 5/31/02.
whitelake
Myst, In regard to new data feed this might have some kind of clue in it.
It is what PCA uses for data feed. Recently I have been having trouble going through Yahoo and have switched over to Wall Street City. It is kind of nice to have a optional back up. This probably doesn't help but it might trigger one of your creative brain cells. Thanks for your response on the data feed.
whitelake
Myst, I see that the user manual says The daily price data will update overnight, so you should check your chart in the morning before the market opens. Is there anyway to get a data feed that will update sometime after the close in the evening. Some people like myself leave early for work. I hit the road at 5:30 a.m. to beat the heat.
whitelake
ACEHI, Your observations seem right on, but what about the quantity of shares? The difference between limit order and actual buy in number of shares I mean. It is good to hear of your experiences with XDev using actual money. Thanks,
whitelake
Myst, OK I see how it shows the limit price now, but what about the number of shares being different. The actual number of shares traded seem to be different from what I would have put on the limit orders following the XDev recommendation? Thanks for being so patient with my questions.
whitelake
Myst, In the following GE run notice that as you walk forward to 3/6/02 and 3/7/02 the recommended sells don't match the actual trades. I know that limit orders can sell at "or better" but that doesn't seem to explain it.
whitelake
Myst, In the chart for G that you posted there are no shares indicated at the next buy of 34.28. Does this mean that there is no need for a limit order at this point as XDev is saying it is waiting for a lower point. If so why is it calling 34.28 the next buy? I see this often and am a little confused.
whitelake
Myst, That is Paint Shop Pro 7.0 and it is good for images for the web, photo enhancement and more than I can utilize. My daughter is the creative one. It is very good for screen captures partial or full. The full version is available for trial for maybe a month.
whitelake
Myst, I was able to squeak out another 7% by opening the bands a little.
whitelake
Myst, The first time wasn't too bad. It would have been a lot easier if my little 1.5 year old grandson wasn't so fond of "bumpaw". Thanks for your help, and I can see you have had a real busy 24 hours. You picked a good time to release XDev as it seems IHUB has light traffic on the weekend.
whitelake
Myst, This is a set up I did for SCH.
This appears to me to be optimized and it is saved. So the question is if I actually wanted to set a limit order for Monday morning what would be the number of shares? I know that it would be 66% of the dollar value of my beginning investment, but how can I make XDev show it. This assumes that Monday would be my first day of starting with SCH.
whitelake
Myst, That will work for me. My CPA son-in-law just got back from a seminar given by K2 Enterprises http://www.k2e.com/ called "EXCEL Beyond the Basics." He brought me the book and CD and he says it is loaded with neat tricks that he didn't know, and he is an Excel whiz. I might get to use some of those new hot tricks in keeping score for XDev.
whitelake
Myst, I have no problem with that. It is the mechanics of keeping track of transactions in XDev going forward that gives me questions. I am used to the PCA program for AIM where you enter the transactions which are usually a little different than what PCA actually asked for. I don't see this capability in XDev. Maybe it is for the 3.0 version. Hope this is "more better" as they say around here.
whitelake
Myst, I am looking at your QQQ example of actual XDev at
http://stockwerld.com/qqq.htm and I have a question. If the buy that you made Friday would have been other than the exact amount called for by XDev how would you have entered it? It is possible that due to problems you would fail to have your limit order in place to make the buy. It looks like the program would see only the $26.31 buy point. How do you adjust if that does not happen? What if it trades at $26.29?
I am having no problems with optimising XDev, but the accounting for the ongoing transactions is giving me a challenge.
whitelake
Larry, The key should come in an email.
whitelake
Myst, I was blessed with an easy install and everything looks fine. MSFT watch out. I don't know if security with PayPal needs a little pad lock icon, but it didn't have one. I went ahead anyway.
whitelake