Problem with age weighted historical simulation (Value at Risk and Expected Shortfall)

Hi,

Im having some problem with my code on age/time-weighted historical simulation when i am trying to calculate VaR and ES on a rolling sample. My calculated VaR is never violated by the actual losses, so there must be something wrong in my code. Any help is appriciated.

Thanks.

data = Nameofdataset
dataTT = table2timetable(data);
head(dataTT,5)
Losses = -data.change; (makes losses positive)
T = size(Nameofdataset);
rT = 500;  (my estimation window)
Lambda = 0.94;
weight = zeros(rT,1);   
alpha = 0.975;
w(rT) = (1-Lambda)/(1-Lambda^rT);
for j=1:rT  
    w(j) = w(rT)*Lambda^(rT-j);
end

for j =rT:1:T 
    rollSample = Losses(1:j,1);
    twLossesRoll = timetable(dataTT.Datum(j-(rT-1):j),Losses(j-(rT-1):j),w,'VariableNames',{'Losses','w'});
    sortedtwLossesRoll = sortrows(twLossesRoll,1,'descend');
    sumW = 0;
    sortedWRoll = sortedtwLossesRoll.w;
    i = 1;
    while sumW<(1-alpha)
        sumW = sumW + sortedWRoll(i);
        i = i + 1;
    end
    twVaRj(j-rT+1,1) = sortedtwLossesRoll.Losses(i-1);
    twES(j-rT+1,1) = mean(sortedtwLossesRoll.Losses(1:i-1));
end

[dpb]

Not much (as in anything) anybody here can do w/ no idea what it is you're trying to do first and without any data.