Burg Extrapolator - MetaTrader 4のためのエキスパート

Updates:

12/26/2008 - corrected the lots calculation function

The EA uses the method of Burg's linear prediction. The linear prediction is based on finding the future values as the values of the lienar functions of the past values. Assume that we have a number of prices x[0]..x[n-1] where the higher index is compliant with the recent price. The prediction of the future price x[n] is calculated as

x[n] = -Sum(a[i]*x[n-i], i=1..p)

where a[i=1..p] - coefficients of the model, p - order of the model. The Burg's method finds the a[] coefficients by decreasing the mean-root-square error on the training last n-p bars.

The inputted data are:

MaxRisk - the maximum risk of all simultaneous deals
ntmax - the maximum number of deals in the same direction
MinProfit - the minimum predicted price the positions should be opened by
MaxLoss - the maximum predicted loss the positions should be closed at
TakeProfit
StopLoss
TrailingStop
PastBars - the number of past bars to be used for future prediction
ModelOrder - the order of the Burg's model as a fraction of the past bars number (0..1)
UseMOM - turns on the detrend of the input data: mom(i)=log[p(i)/p(i-1)]
UseROC - turns on the detrend of the input data: roc=100*(p(i)/p(i-1)-1)

Only one of the variables UseMOM and UseROC can have the true value simultaneously, i.e. UseMOM=true and UseROC=true are not allowed.

As the most of optimized EAs, Burg Extrapolator work well only on the training bars. The EA will steadily lose without a constant reoptimization.