Time series forecasting using regression learner

Dear All,

I am deploying AI to predict covid19 curve flaten.

I have used regression learner to predict confirmed cases; one dimension series (t_series).

The exported regression model shows good convergence.

However, the prediction result is not good.

I tried to preprocess data (using mean / standard deviation) but yet no improvement.

%% Preprocessing
numTimeStepsTrain = floor(0.95*numel(t_series)) ;% take 95% of data fro training
dataTrain = t_series(1:numTimeStepsTrain+1);
dataTest = t_series(numTimeStepsTrain+1:end);

%% Standardize Data
mu = mean(dataTrain);
sig = std(dataTrain);
dataTrainStandardized = (dataTrain - mu) / sig;

%% Prepare Predictors and Responses
XTrain = dataTrainStandardized(1:end-1);
YTrain = dataTrainStandardized(2:end);

%% call leaner regression app (from APP menu)

I tried Generalized linear model regression function (glmfit) but also nothing improved.

%% call leaner regression app OR Generalized linear model regression function (glmfit)
mdl = glmfit(XTrain,YTrain);
yfit = glmval(mdl,XTrain,'probit','size',numel(YTrain));
plot(XTrain, YTrain,'o',XTrain,yfit,'-','LineWidth',2)

% results are diverging 

Thank you for your kind help.