Thread Subject:

generate 50x1000 matrix given variance and mean

Ok so I know how to generate a random matrix of 50x1000. What I need to do now is create it with mean 20 and variance 100.

All I have is x = randn(50,1000);

but the distribution has to have mean 20 and variance 100?? little hint here please?

On Mar 24, 7:39=A0pm, "Jose Valerio" <lape...@gmail.com> wrote:
> Ok so I know how to generate a random matrix of 50x1000. What I need to d=
o now is create it with mean 20 and variance 100.
>
> All I have is x =3D randn(50,1000);
>
> but the distribution has to have mean 20 and variance 100?? little hint h=
ere please?


Hint:
   z =3D randn(1000, 1) * sigma + mu;

This creates a normal random vector with mean mu and variance sigma^2.

Andy

andywocky@gmail.com wrote in message <d0d61c92-546f-42c9-af68-f1a5990fe29f@p6g2000pre.googlegroups.com>...
> On Mar 24, 7:39=A0pm, "Jose Valerio" <lape...@gmail.com> wrote:
> > Ok so I know how to generate a random matrix of 50x1000. What I need to d=
> o now is create it with mean 20 and variance 100.
> >
> > All I have is x =3D randn(50,1000);
> >
> > but the distribution has to have mean 20 and variance 100?? little hint h=
> ere please?
>
>
> Hint:
> z =3D randn(1000, 1) * sigma + mu;
>
> This creates a normal random vector with mean mu and variance sigma^2.
>
> Andy

I found this on matlab help:
Generate a random distribution with a specific mean and variance . To do this, multiply the output of randn by the standard deviation , and then add the desired mean. For example, to generate a 5-by-5 array of random numbers with a mean of .6 that are distributed with a variance of 0.1,

x = .6 + sqrt(0.1) * randn(5)

So I did
x = 20 + sqrt(100) * randn(50,1000)

"Jose Valerio" <lapenda@gmail.com> wrote in message <gqbqul$dgj$1@fred.mathworks.com>...
> Ok so I know how to generate a random matrix of 50x1000. What I need to do now is create it with mean 20 and variance 100.
>
> All I have is x = randn(50,1000);
>
> but the distribution has to have mean 20 and variance 100?? little hint here please?

Read this:

doc randn

HTH,
John

"John D'Errico" <woodchips@rochester.rr.com> wrote in message <gqbshb$6lu$1@fred.mathworks.com>...
> "Jose Valerio" <lapenda@gmail.com> wrote in message <gqbqul$dgj$1@fred.mathworks.com>...
> > Ok so I know how to generate a random matrix of 50x1000. What I need to do now is create it with mean 20 and variance 100.
> >
> > All I have is x = randn(50,1000);
> >
> > but the distribution has to have mean 20 and variance 100?? little hint here please?
>
> Read this:
>
> doc randn
>
> HTH,
> John

I think the link didn't come up??
I'm not 100% sure if the way I did it is right

"Jose Valerio" <lapenda@gmail.com> wrote in message <gqbssh$gfj$1@fred.mathworks.com>...
> "John D'Errico" <woodchips@rochester.rr.com> wrote in message <gqbshb$6lu$1@fred.mathworks.com>...
> > "Jose Valerio" <lapenda@gmail.com> wrote in message <gqbqul$dgj$1@fred.mathworks.com>...
> > > Ok so I know how to generate a random matrix of 50x1000. What I need to do now is create it with mean 20 and variance 100.
> > >
> > > All I have is x = randn(50,1000);
> > >
> > > but the distribution has to have mean 20 and variance 100?? little hint here please?
> >
> > Read this:
> >
> > doc randn
> >
> > HTH,
> > John
>
> I think the link didn't come up??
> I'm not 100% sure if the way I did it is right

I think I got it now.
r = 20 + 100.*randn(50,1000);

"Jose Valerio" <lapenda@gmail.com> wrote in message <gqbtdm$3f1$1@fred.mathworks.com>...

> > I'm not 100% sure if the way I did it is right
>
> I think I got it now.
> r = 20 + 100.*randn(50,1000);

Very, very close. You asked for the VARIANCE to
be 100.

Multiply a unit normal random variable by a constant
value k, and the standard deviation will be k. The
variance (the square of the standard deviation) will
be k^2.

So you need to multiply by sqrt(100).

John

"John D'Errico" <woodchips@rochester.rr.com> wrote in message <gqbu3t$p8f$1@fred.mathworks.com>...
> "Jose Valerio" <lapenda@gmail.com> wrote in message <gqbtdm$3f1$1@fred.mathworks.com>...
>
> > > I'm not 100% sure if the way I did it is right
> >
> > I think I got it now.
> > r = 20 + 100.*randn(50,1000);
>
> Very, very close. You asked for the VARIANCE to
> be 100.
>
> Multiply a unit normal random variable by a constant
> value k, and the standard deviation will be k. The
> variance (the square of the standard deviation) will
> be k^2.
>
> So you need to multiply by sqrt(100).
>
> John

isn't that what I had on my second post? x = 20 + sqrt(100) * randn(50,1000)

"Jose Valerio" <lapenda@gmail.com> wrote in message <gqbv00$4v4$1@fred.mathworks.com>...
> "John D'Errico" <woodchips@rochester.rr.com> wrote in message <gqbu3t$p8f$1@fred.mathworks.com>...
> > "Jose Valerio" <lapenda@gmail.com> wrote in message <gqbtdm$3f1$1@fred.mathworks.com>...
> >
> > > > I'm not 100% sure if the way I did it is right
> > >
> > > I think I got it now.
> > > r = 20 + 100.*randn(50,1000);
> >
> > Very, very close. You asked for the VARIANCE to
> > be 100.
> >
> > Multiply a unit normal random variable by a constant
> > value k, and the standard deviation will be k. The
> > variance (the square of the standard deviation) will
> > be k^2.
> >
> > So you need to multiply by sqrt(100).
> >
> > John
>
> isn't that what I had on my second post? x = 20 + sqrt(100) * randn(50,1000)

Yes. This is now correct. So then why did you
change it incorrectly in your penultimate post?

John

"Jose Valerio" <lapenda@gmail.com> wrote in message <gqbssh$gfj$1@fred.mathworks.com>...
> "John D'Errico" <woodchips@rochester.rr.com> wrote in message <gqbshb$6lu$1@fred.mathworks.com>...
> > "Jose Valerio" <lapenda@gmail.com> wrote in message <gqbqul$dgj$1@fred.mathworks.com>...
> > > Ok so I know how to generate a random matrix of 50x1000. What I need to do now is create it with mean 20 and variance 100.
> > >
> > > All I have is x = randn(50,1000);
> > >
> > > but the distribution has to have mean 20 and variance 100?? little hint here please?
> >
> > Read this:
> >
> > doc randn
> >
> > HTH,
> > John
>
> I think the link didn't come up??
> I'm not 100% sure if the way I did it is right

I think I got it now.
r = 20 + 100.*randn(50,1000);

"John D'Errico" <woodchips@rochester.rr.com> wrote in message <gqbvda$1ob$1@fred.mathworks.com>...
> "Jose Valerio" <lapenda@gmail.com> wrote in message <gqbv00$4v4$1@fred.mathworks.com>...
> > "John D'Errico" <woodchips@rochester.rr.com> wrote in message <gqbu3t$p8f$1@fred.mathworks.com>...
> > > "Jose Valerio" <lapenda@gmail.com> wrote in message <gqbtdm$3f1$1@fred.mathworks.com>...
> > >
> > > > > I'm not 100% sure if the way I did it is right
> > > >
> > > > I think I got it now.
> > > > r = 20 + 100.*randn(50,1000);
> > >
> > > Very, very close. You asked for the VARIANCE to
> > > be 100.
> > >
> > > Multiply a unit normal random variable by a constant
> > > value k, and the standard deviation will be k. The
> > > variance (the square of the standard deviation) will
> > > be k^2.
> > >
> > > So you need to multiply by sqrt(100).
> > >
> > > John
> >
> > isn't that what I had on my second post? x = 20 + sqrt(100) * randn(50,1000)
>
> Yes. This is now correct. So then why did you
> change it incorrectly in your penultimate post?
>
> John

Because I was doing it from this:
Generate values from a normal distribution with mean 1 and standard deviation 2.
r = 1 + 2.*randn(100,1);

and I miss-read it, it's doing it for std 2 not VARIANCE, that's where I made the mistake. Thanks for your help