MATLAB Answers

How to add confidence intervals (95%) in a return level plot for a Weibull3 distribution?

18 views (last 30 days)
Mari Sandvik
Mari Sandvik on 24 Oct 2017
Commented: Mari Sandvik on 3 Nov 2017
I have gotten a code from a colleague, which I have to update. The code (EVA_Weibull3.m) conducts a extreme value analysis by using the Weibull3 distribution for wind data (Wind_Krakenes_2005-2016). To evaluate the max values a block maximum method is used, with a block width of 30 days.
The script is working, but I want to include confidence intervals for the Weibull 3 distribution when plotting the return level. My figure now only contains the block maximum value of the data (blue dots) and the return value of the created distribution (black line).
My problem is that I'm not sure how to calculate the 95% confidence intervals?
I have attached the code, since it is almost 200 lines and I'm not sure what information is needed to calculate the 95% confidence intervals.
Mari

  0 Comments

Sign in to comment.

Answers (1)

Mary Abbott
Mary Abbott on 30 Oct 2017
Hello,
The "wblfit" function can be used to estimate the confidence interval of the parameters of a Weibull distribution:
https://www.mathworks.com/help/stats/wblfit.html

  1 Comment

Mari Sandvik
Mari Sandvik on 3 Nov 2017
Hi,
The wblfit function does estimate the confidence interval of a 2 parameter Weibull distribution, where the location file is equal zero. However I want to know how to calculate the confidence interval of a 3 parameter Weibull distribution.
If I had decided to only do a 2 parameter Weibull distribution, how would you plot these confidence intervals in a return level plot as shown above.
Mari

Sign in to comment.

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by