Anyone????
|
Anyone????
pm sent, thanks.
The data for my investing platform is in an Excel file at the moment...do I need to convert it to another format?
Thank you in advance for any advise!
hello guy and gals.
I'm experimenting with option and try to download the data to excel spreadsheet , the problem is that there is hundred of lines and data and the spreadsheet stuck all the time , is there better way to download the data to other sheet or product and do the math there?
thanks.
Hi guys,
I need to construct a scatter graph in Excel of the "Expected Profit" (as outputted by the MT4 Strategy Tester) versus a variable that I ran an optimization on called "Cutoff." My aim is to use Excel to generate a best-fit curve for the resulting graph and incorporate the function in one of my EAs.
When I tried doing this I ran into a problem: Some optimization cycles yield very few trades (ex. 2), while others yield magnitudes more (ex. 450). It makes no sense to plot the "Expected Profits" from the 2-trade-optimization on the same graph as the 450-trade one; but I can't just toss those cycles out either.
I need to somehow apply a "weight" to the various "Expected Profits" variables based on the number of trades. This is what I cannot figure out.
I attempted to make a "Weighted Expected Profits" column and simply use that in my graph in place of "Expected Profits," but the results were not what I intended; I must have done it incorrectly.
Here is an example:
![]()
you are totally talking about Datamining
or regression analysis
I doubt your Excel (2003 2007 2010) could do something that advanced
I learnt datamining in my master course, if I pay handsomely by you , I might able to do it for you -- I have not yet look into your data in details -- to see whether it is within my ability
Are you certain that it is as difficult as you say? The main part for me is not generating the function; that is secondary. For now I just want to graph the scatter chart properly.
I am not sure how to do it in excel
you could use some automatic estimate function, but it is very primitive
===
I am quite certain that automatic programming -- datamining -- we seldom need to do real programming in those datamining software
it will find the weighted and variable to fit the best 91% of your best outcome (in this case, you find the few best EA that are able to get you highest profit and reliable consistency)
so I know a lot about datamining (my excel knowledge limit to pivot table , VBA etc) to get the best fitted curve or derive the formula for you
I think I could be in the consultant level in datamining to gave you my previous REPLY
but if I do it, it will come with a price tag
I understand what datamining is, and I am pretty certain that what I want is not datamining at all. From what I understand, datamining is applying various mathamatical/statistical functions to an (often large) set of variables and using these variables in combination to draw a conclusion out of the 'cloud' of data.
My situation is totally different. I have only 1 variable and nothing to 'mine.' All I need is a coefficient to multiply the "Expected Payoff" by in a way that would compensate for the chart-skewing effect caused by throwing a 2-trade optimization run into the mix with 250-trade runs.
I am certain that there is a very simply solution.... such as dividing the Expected Payoff by the sum of all trade-amounts from all cycles.... or something like that.
Last edited by mfurlend; 06-27-2010 at 01:43 AM.
Hottest Threads
view more |
||||||||||
Most Popular Threads
view more |
There are currently 2 users browsing this thread. (0 members and 2 guests)
Bookmarks