Increasing Accuracy of Univariate NARNET Time series

Please I ran this code for a univariate prediction but the seasonality in my result has a wide gap between the actual and predicted. What can I do to improve the accuracy. Love to hear from you.

clear;clc;

Xall= xlsread('APB.xlsx','Autumn','b2:b277');
Y=Xall(1:256,1);
% X=cell2mat(X);
Xtest= Xall(257:276,1);
% T = tonndata(X,false,false);

net = narnet(1:3,25); % creates the autoregressive NN
net.trainParam.showWindow = true; % shows the training window

T = tonndata(Y,false,false); % converts the data to a ANN cell array form.
[Xs,Xi,Ai,Ts] = preparets(net,{},{},T); % 
net = closeloop(train(net,Xs,Ts,Xi,Ai));

Ypred = nan(23,1);
% tre=Ypred;
Ypred = tonndata(Ypred,false,false);
Ypred(1:4) = T(end-3:end);

[xc,xic,aic,tc] = preparets(net,{},{},Ypred);
Ypred = fromnndata(net(xc,xic,aic),true,false,false);
save('netAPB_Autumn')
% Performance= perform(net,tc,Ypred)

xlswrite('APB Predicted Result.xlsx', Ypred, 'a2:a20')
xd=1:1:20;
figure, plot(xd,Ypred,'--*', xd,Xtest, '-.o')
        xlabel('Days')
        ylabel('Prediction')
        legend('Predicted', 'Actual')