To obtain p-values for the parameter estimates in ARIMA model estimation, you can use the 'infer' function. Here's an example:
Mdl1 = estimate(Mdl, y);
[~, ~, ~, Cov] = infer(Mdl1, y);
SE = sqrt(diag(Cov));
tStat = Mdl1.Coefficients.Estimate ./ SE;
% Compute the p-values using the cumulative distribution function (CDF) of the t-distribution
df = length(y) - numel(Mdl1.Coefficients.Estimate);
pValues = 2 * (1 - tcdf(abs(tStat), df));
Thanks,
Balaji