Random Number Sampling Between a Specific Range of a Non-Uniform Distribution

I need to sample from a Gaussian mixture distribution (GMmodel) using Monte Carlo method. For certain scenarios values need to be sampled from a specific range as determined by the x-axis (B). Currently I'm attempting to use logical indexing to do so, however I'm either missing something or have made an error as so far the values I get at the end don't make sense - most values are negative, which is impossible given the system I'm attempting to model.

Here is the relevant code:

load('T.mat');

load('B.mat');

t = 'baseline SPRING'

for i = 1:Mcrun;

r = GMmodel.random(1);

if t == 'baseline SPRING'

 LOGIC=r(:,1)<=10.57&r(:,1)>0;

elseif t == 'baseline SUMMER'

 LOGIC=r(:,1)<=35&r(:,1)>0;

Bm=r(:,1).*LOGIC;

Tm=r(:,2).*LOGIC;

From here I plug Bm and Tm into equations. I've traced back the error (negative values) here.

Would using logical indexing the best way to do this type of sampling? Is there a better method?

I'm new to MatLab and would greatly appreciate any guidance.

[Torsten]

You forgot the "end"-statement after the line "LOGIC=r(:,1)<=35&r(:,1)>0;".

Best wishes

Torsten.