Thread Subject: Help with logic. Making two vectors equal in length.

Dear All,
I have three vectors vector1 and vector 2 and finalvector2, such
that:

% Generating vector 1 %%%%

mu1 = 9.5;
sd1 = 0.645;

feed1 = normrnd(mu1,sd1,1,1000000);

vector1 = cumsum(feed1);

%cutoff
cutoff = max(vector1);


% Generating vector 2%%%%%

mu2 = 19.7; %mu2 needs to be larger than mu1. It cannot be
same or close to the value of mu1
sd2 = 0.645;

feed2 = normrnd(mu2,sd2,1,1000000); %length of samples here is
1000000

vector2 = cumsum(feed2);

%final vector2

finalvector2 = vector2 (vector2 < cutoff);

end


Now it is obvious, that the number of elements in finalvector2 is
going to be lot less than vector1, after application of 'cutoff'.
Ofcourse vector2 and vector 1 will be same in size. But what I want
is, even after the cutoff application, I want the number of elements
in finalvector2 to be more or less same as the number of elements in
vector1.


It is clear that I need to generate more elements in feed2 to achieve
this. The query is 'by how much?'

I would have guessed it would be a relation between the following
'cutoff' and 'No of elements I require' and 'Number of extra elements
in feed to generate'

I am doing something wrong in calculating this relationship and
implementing it though....

Can somebody please help me?

Will be very grateful for your responses.

Cheers,
Adshak.

On Jul 18, 12:15 pm, Adshak <adshaikh.hip...@googlemail.com> wrote:
> Dear All,
> I have three vectors vector1 and vector 2 and finalvector2, such
> that:
>
> % Generating vector 1 %%%%
>
> mu1 = 9.5;
> sd1 = 0.645;
>
> feed1 = normrnd(mu1,sd1,1,1000000);
>
> vector1 = cumsum(feed1);
>
> %cutoff
> cutoff = max(vector1);
>
> % Generating vector 2%%%%%
>
> mu2 = 19.7;           %mu2 needs to be larger than mu1. It cannot be
> same or close to the value of mu1
> sd2 = 0.645;
>
> feed2 = normrnd(mu2,sd2,1,1000000); %length of samples here is
> 1000000
>
> vector2 = cumsum(feed2);
>
> %final vector2
>
> finalvector2 = vector2 (vector2 < cutoff);
>
> end
>
> Now it is obvious, that the number of elements in finalvector2 is
> going to be lot less than vector1, after application of 'cutoff'.
> Ofcourse vector2 and vector 1 will be same in size. But what I want
> is, even after the cutoff application, I want the number of elements
> in finalvector2 to be more or less same as the number of elements in
> vector1.
>
> It is clear that I need to generate more elements in feed2 to achieve
> this. The query is 'by how much?'
>
> I would have guessed it would be a relation between the following
> 'cutoff' and 'No of elements I require' and 'Number of extra elements
> in feed to generate'
>
> I am doing something wrong in calculating this relationship and
> implementing it though....
>
> Can somebody please help me?
>
> Will be very grateful for your responses.
>
> Cheers,
> Adshak.

Dear ALL,

To further clarify my query, how can I generate 1 million samples of
feed2 = (mu2,sd2,1,1000000) with the condition max(vector2)<= max
(vector1) being satisfied.

I will be very grateful for any response that can help me.

sincere thanks in advance.

But you haven't said what values you want the "extra" elements to
take. In this example, I set them either to zero or cutoff. See if
this is what you're looking for:

clc;
close all;
workspace;
% Generate values from a normal distribution with specified mean and
standard deviation.
mu1 = 9.5;
sd1 = 0.645;
feed1 = mu1 + sd1 * randn(100, 1);
vector1 = cumsum(feed1);
%cutoff
cutoff = max(vector1);
% Generating vector 2%%%%%
mu2 = 19.7; %mu2 needs to be larger than mu1. It cannot be
same or close to the value of mu1
sd2 = 0.645;
feed2 = mu2 + sd2 * randn(100, 1);
vector2 = cumsum(feed2);
%final vector2
subplot(2,2,1);
plot(feed1, 'r');
hold on;
plot(feed2, 'g');
subplot(2,2,2);
plot(vector1, 'r');
hold on;
plot(vector2, 'g');
finalvector2a = cutoff * ones(length(vector2), 1);
finalvector2a(vector2 < cutoff) = vector2(vector2 < cutoff);
subplot(2,2,3);
plot(finalvector2a, 'b');
finalvector2b = cutoff * zeros(length(vector2), 1);
finalvector2b(vector2 < cutoff) = vector2(vector2 < cutoff);
subplot(2,2,4);
plot(finalvector2b, 'b');

Adshak <adshaikh.hipnet@googlemail.com> wrote in message <730987aa-f3c1-45d6-a612-d8e4b3877332@18g2000yqa.googlegroups.com>...
>
> To further clarify my query, how can I generate 1 million samples of
> feed2 = (mu2,sd2,1,1000000) with the condition max(vector2)<= max
> (vector1) being satisfied.
>
> I will be very grateful for any response that can help me.
>

This does what you want:

http://www.mathworks.com/matlabcentral/fileexchange/23832

Bruno