Exceedance probability of normal and logarithmic distribution

I have estimated the parameters i.e. mean and standard deviation of a normal and lognormal distribution for my 30 years river discharge data using command NORMFIT and LOGNFIT. Now I want to determine the magnitude of peak discharge value that will be exceeded with a probability of 0.01(the 100-year flood) and the probability of exceeding of mean discharge.

If anyone give me the command that I can use in Matlab to solve this problem, it will be really helpful for me.

Answer 1

[the cyclist]

norminv and logninv

[Shumon]

Thanks. Can you tell me, how can I find out the probability which will exceed mean?

[the cyclist]

For that you will need the normcdf and logncdf commands.

For example, if your parameters are

mu    = 3;
sigma = 4;

then probability of being less than a value x is

p_x = logncdf(x,mu,sigma),

so the probability of being less than the mean is

p_mu = logncdf(mu,mu,sigma),

and the probability of exceeding that is

1 - p_mu

Of course the probability of exceeding the mean of the normal is 1/2. (You will get that if you calculate it analogously to the above.)

[See Kai Xiang]

Hi, it has been awhile since this question was asked, but my question was not answered so i hope this would bring me some enlightenment. In the logncdf case it is possible to plot a cdf graph. However, if i wanted to plot the 1-cdf graph, 1-p_mu will not work? May i know then how do i plot the 1-cdf graph?