Interval Widths and MCMC Sampling

Hi,

In the package documentation, it is mentioned that "By default Prophet will only return uncertainty in the trend and observation noise. To get uncertainty in seasonality, you must do full Bayesian sampling.".

Since I want the true confidence intervals and treat them as anomaly thresholds, I tried specifying the MCMC parameter = 300. When I plot the components, I do see intervals around the fitted values now but when I get lower and upper bounds (specifying an interval width of 0.997) on forecasted datapoints, there doesn't seem to be any differences when I compare values of bounds obtained with or without MCMC sampling.

Is the uncertainty in seasonality components accounted for in the lower/upper bounds when MCMC parameter > 0?