• We are pleased to announce that the winner of our Feedback Prize Draw for the Winter 2024-25 session and winning £150 of gift vouchers is Zhao Liang Tay. Congratulations to Zhao Liang. If you fancy winning £150 worth of gift vouchers (from a major UK store) for the Summer 2025 exam sitting for just a few minutes of your time throughout the session, please see our website at https://www.acted.co.uk/further-info.html?pat=feedback#feedback-prize for more information on how you can make sure your name is included in the draw at the end of the session.
  • Please be advised that the SP1, SP5 and SP7 X1 deadline is the 14th July and not the 17th June as first stated. Please accept out apologies for any confusion caused.

April 2014 question 4

E

Esnath

Member
Anyone know how they drew the GEV and GPD distribution graphs for the flood data?

In the April 2014 past paper
 
Hello

The blue bar charts are frequency graphs.

We start by subdividing the data into blocks of even size, eg 5 years.
Then find the maxima for each block. These are:
128,000, 230,000, 122,000, 84,400, 64,300, 185,000, 85,400, 185,000, 110,000, 92,100, 203,000 and 135,000.

Then we produce a frequency chart:

Flood size Number of block maxima in this range
60-95K 4
95-130K 3
130-165K 1
165-200K 2
200-235K 2

So we have 12 block maxima observation and these are split into bars of height (ratio) 4:3:1:2:2 on the graph.

If we are comparing this frequency distribution chart with a PDF of a distribution, we want the area under the frequency distribution chart to sum to 1. Take the 130K to 165K block where there is a single observation. The area of this block must be 1/12. The width is 35K so the height will be (1/12) divided by 35,000 = 0.00000238. The other blocks will be multiples of this.

The PDF is fitted using MLE with a GEV distribution PDF.

For the other graph, using threshold exceedances, the approach will be similar.

Choose a threshold, eg 100,000.
Look at all the flood levels above 100,000 (can use the ordered data for this).
Work out the exceedance in each case, eg 2000, 2000, 8000, 10,000, 13,000 etc.
Do a frequency table similar to the one above.
Plot the blue graph.

Does this help?
Anna
 
Back
Top