It would be significantly easier to offer assistance if you posted your code, or some simplified test case that reproduced what you are seeing.
Mike --- In [email protected], Keith McCombs <kmcco...@...> wrote: > > Mike -- > This is just a guess. But, each bank probably has a different total > number of bars. So the Cum() is going to be 'acCumulated' over a > different period. Rather than using Cum() why not use the sum( ARRAY, > periods ). And make sure the 'periods' does not exceed the minimum > number of bars for all the stocks used in the group that you are analyzing. > -- Keith > > On 3/19/2010 16:32, Mike wrote: > > > > I'm new to this add to composite function and I am using this code to > > plot an AD line. > > ............................................... > > int_string = "Banking"; //MarketID(1) > > Difference = ( Foreign("~adv_" +int_string, "C") - Foreign("~dec_" > > +int_string, "C") )/ ( Foreign("~unc_" +int_string, "C")+ 1 ); > > DiffSqrt = IIf( Difference > 0, sqrt( Difference ), - sqrt( - > > Difference ) ); > > BankADLine = Cum( DiffSqrt ); > > _TRACE("BankADLine "+ BankADLine ); > > ............................................... > > > > I scaned a watchlist of Banking stocks to get the > > ~adv_Banking,~dec_Banking, and ~unc_Banking composites which are in > > group 253. > > Whenever I switch chart tabs at the top of the chart display and the > > ticker changes I get a different trace value for the BankADLine. > > > > How are the composite symbols changing their scanned values. I've even > > tried the setforeign("sym"); and RestorePriceArrays() ; to lock in the > > symbol I think the code is working with but they are completely ignored. > > > > I'm trying to get the AD line values for multiple watchlists/sectors > > in one indicator. So I want to repeat "this" code multiple times. > > > > Please help, my head is starting to hurt...this desk is really hard! > > > > Thank you for any assistance! > > > > mike > > > > >
