Apparently the cconv function is not yet implemented for Octave's signal package. What is a simple way to reproduce its functionality?
The code which would like to execute is:
pkg load signal
n=linspace(0,1,1024);
s=sin(2.*pi.*1.*n)+.3.*sin(2.*pi.*4.*n);
x=s+.1.*randn(size(s));
L=100;
y=cconv(x,ones(1,L)./L,1024);
figure,plot(y);
xlabel('n'),ylabel('y[n]'),title('Output y[n]')
But the following error message is shown after trying to use cconv():
warning: the 'cconv' function belongs to the signal package from Octave Forge but
has not yet been implemented.
The code is an assignment provided by edX' "Discrete Time Signals and Systems" where an integrated Matlab-interface is provided. I would like to use Octave though.
function a = cconv( b, c )
nb = length(b) ;
nc = length(c) ;
##
## Ensure the two vectors are the same length
##
if (nb < nc)
b = [ b zeros(1,nc-nb) ] ;
endif
if (nc < nb)
c = [ c zeros(1,nb-nc) ] ;
endif
a = ifft(fft(b) .* fft(c)) ;
##
## Get rid of any tiny imaginary bits that should be zero
##
if (all(b == real(b)) && all(c == real(c)))
a = real(a) ;
endif
endfunction
(source)
What about this implimentation from File Exchange?
http://www.mathworks.com/matlabcentral/fileexchange/13030-circular-convolution/content/cconv.m
Here's how to do circular convolution that gives identical results to the ifft(fft.*fft) method.
x=[1,2,3,4,3,2,5,0,1,-10,-1,3];
h=[0.5, 0.3, 0.15, 0.05];
Lx=length(x);
Lh=length(h);
y=zeros(1,Lx+Lh-1);
% circular convolution without zero padding and fft
y=conv([x((end-Lh+2):end) x],h);
y1=y(Lh:end-Lh+1)
% fft method for circular convolution (must zero pad to equal size)
y2=ifft(fft(x).*fft([h zeros(1,Lx-Lh)]))
Related
I was trying to solve a system of non-linear equations and check the stability of its equilibrium points. Initially, I declared two equations
adot = -a + 2*a^3 + b
bdot = -a -b
By equating both the equations with 0, I get the equilibrium points. Now, I am trying to get the Jacobian of the [adot;bdot] matrix using the jacobian([adot;bdot],[a,b]) method of symbolic package in Octave, which should just return a matrix whose items are the partial derivatives of "adot" and "bdot" w.r.t. "a" and "b", but it gives the following error
error: subscript indices must be integers or boolean
Can anyone tell me where I'm going wrong with this?
Edit: I'm adding the complete code down below:
pkg load symbolic
syms x y
xdot = -x + 2*x^3 + y;
ydot = -x - y;
[xeq,yeq] = solve(xdot==0,ydot==0);
xeq = double(xeq);
yeq = double(yeq);
jacobian_matrix = jacobian([xdot;ydot]);
At this point I'm getting the above mentioned error. The values in the [xeq,yeq] matrix are the equilibrium points of the system, which are to be used later on.
here is my problem. I'm trying to solve a system of two differential equations thanks to the two functions below. The part of the code that give me some trouble is the variable "rho". "rho" is a function which values are given from a file and that I tried to put in the the variable rho.
function [xdot]=f2(x,t)
# Parameters of the equations
t=[1:1:35926];
x = dlmread('data_txt.txt');
rho=x(:,4);
beta = 0.68*10^-2;
g = 1.5;
l = 1.6;
# Definition of the system of 2 ODE's :
xdot(1) = ((rho-beta)/g)*x(1) + l*x(2);
xdot(2) = (beta/g)*x(1)-l*x(2);
endfunction
.
# Initial conditions for the two variables :
x0 = [0;1];
# Definition of the time-vector -> (initial time,temporal mesh,final time) :
t = linspace (1, 10, 10000);
# Resolution with the lsode routine :
x = lsode ("f2", x0, t);
# Plot of the two curves :
plot (t,x);
When I run my code, I get the following error:
>> resolution_effective2
error: f2: A(I) = X: X must have the same size as I
error: called from
f2 at line 34 column 9
resolution_effective2 at line 8 column 3
error: lsode: evaluation of user-supplied function failed
error: called from
resolution_effective2 at line 8 column 3
error: lsode: inconsistent sizes for state and derivative vectors
error: called from
resolution_effective2 at line 8 column 3
I know that my error comes from a mismatch of size between some variable but I have been looking for the error for days now and I don't see. Could someone try to give and explain me an effective correction ?
Thank you
The error might come from your function f2. Its first argument x should have the same dimension as x0 since x0 is the initial condition of x.
In your case, whatever you intent to be the first argument of f2 is ignored since in your function you do x = dlmread('data_txt.txt'); this seems to be a mistake.
Then, xdot(1) = ((rho-beta)/g)*x(1) + l*x(2); will be a problem since rho is a vector.
You need to check the dimensions of x and rho.
32-bit Octave has a limit on the maximum number of elements in an array. I have recompiled from source (following the script at https://github.com/calaba/octave-3.8.2-enable-64-ubuntu-14.04 ), and now have 64-bit indexing.
Nevertheless, when I attempt to perform elementwise multiplication using a broadcast function, I get error: out of memory or dimension too large for Octave's index type
Is this a bug, or an undocumented feature? If it's a bug, does anyone have a reasonably efficient workaround?
Minimal code to reproduce the problem:
function indexerror();
% both of these are formed without error
% a = zeros (2^32, 1, 'int8');
% b = zeros (1024*1024*1024*3, 1, 'int8');
% sizemax % returns 9223372036854775806
nnz = 1000 % number of non-zero elements
rowmax = 250000
colmax = 100000
irow = zeros(1,nnz);
icol = zeros(1,nnz);
for ind =1:nnz
irow(ind) = round(rowmax/nnz*ind);
icol(ind) = round(colmax/nnz*ind);
end
sparseMat = sparse(irow,icol,1,rowmax,colmax);
% column vector to be broadcast
broad = 1:rowmax;
broad = broad(:);
% this gives "dimension too large" error
toobig = bsxfun(#times,sparseMat,broad);
% so does this
toobig2 = sparse(repmat(broad,1,size(sparseMat,2)));
mult = sparse( sparseMat .* toobig2 ); % never made it this far
end
EDIT:
Well, I have an inefficient workaround. It's slower than using bsxfun by a factor of 3 or so (depending on the details), but it's better than having to sort through the error in the libraries. Hope someone finds this useful some day.
% loop over rows, instead of using bsxfun
mult_loop = sparse([],[],[],rowmax,colmax);
for ind =1:length(broad);
mult_loop(ind,:) = broad(ind) * sparseMat(ind,:);
end
The unfortunate answer is that yes, this is a bug. Apparently #bsxfun and repmat are returning full matrices rather than sparse. Bug has been filed here:
http://savannah.gnu.org/bugs/index.php?47175
How looks the syntax of a regression with a quadratic term and interaction in SPSS? In R the code would be:
fit <- lm(c ~ a*b + a*I(b^2), dat)
or
fit <- lm(c ~ a*(b+I(b^2), dat)
Thanks for help.
Using REGRESSION you need to actually make the variables in the SPSS data file before submitting the command. So if your variables were named the same:
COMPUTE ab = a*b. /*Interaction*/.
COMPUTE bsq = b**2. /*squared term*/.
COMPUTE absq = a*bsq. /*Interaction with squared term*/.
Then these can be placed on the right hand side of your regression equation.
REGRESSION VARIABLES=a,b,bsq,absq,c
/DEPENDENT=c
/METHOD=ENTER a,b,bsq,absq.
I thought you could only do factor variables for the interactions - but I was wrong, you can do continuous variables as well (sorry!). Here is an example using MIXED (still you need to make the seperate variables if using REGRESSION).
INPUT PROGRAM.
LOOP Case = 1 TO 200000.
END CASE.
END LOOP.
END FILE.
END INPUT PROGRAM.
COMPUTE a = RV.BERNOULLI(0.5).
COMPUTE b = RV.NORMAL(0,1).
COMPUTE ab = a*b /*Interaction*/.
COMPUTE bsq = b**2 /*squared term*/.
COMPUTE absq = a*bsq /*Interaction with squared term*/.
COMPUTE c = 0.5 + 0.2*a + 0.1*b -0.05*ab + .03*bsq -.001*absq + RV.NORMAL(0,1).
VARIABLE LEVEL a (NOMINAL).
RECODE a (0 = 2)(ELSE = COPY).
MIXED c BY a WITH b bsq
/FIXED = a b b*b a*b
/PRINT SOLUTION.
I am new to Octave although I can say I am an expert Matlab user. I am running Octave on a Linux server (Red Hat) remotely through PuTTY, from a windows machine.
I am observing a very strange behavior in Octave. I call myfun(a) which performs as expected giving the sought results. Now, if I run, say, myfun(b) with b!=a, I get again myfun(a). Clear -f does not solve the problem. I need to reboot octave to change the parameters.
What am I doing wrong?
Thanks a lot
Francesco
This is the code for the function I mentioned:
function [a, v, obj, infos, iter] = mle_garch( p )
#{
% this function estimates the GARCH(1,1) parameters
% it is assumed we pass the adjusted price level p
#}
global y = (diff(log(p))-mean(diff(log(p))))*100;
global h = zeros(size(y));
a0 = [var(y)*0.9; 0.8; 0.1];
[a, obj, infos, iter] = sqp(a0, #loglike_garch, [], #loglike_con, [], [], 1000);
v = sqrt(h * 260);
endfunction
function g = loglike_garch( a )
global y h
n = length(y);
h(1) = var(y);
for i = 2 : n,
h(i) = a(1) + a(2) * h(i-1) + a(3) * y(i-1)^2;
endfor
g = 0.5 * ( sum(log(h)) + sum(y.^2./h) ) / n;
endfunction
function f = loglike_con( a )
f = [1;0;0;0] + [0 -1 -1;eye(3)] * a;
endfunction
I'm assuming the myfun you mentioned is mle_garch. The problem is the way you're initializing the global h and v variables (do you really need them to be global?). When you have a piece of code like this
global y = (diff(log(p))-mean(diff(log(p))))*100;
global h = zeros(size(y));
the values of y and h are defined the first time only. You can change their values later on, but this specific lines will never be ran again. Since your code only uses the input argument to define these two variables, the value which you use to run the function the first time will be used every single other time. If you really want to keep those variables global, replace it with the following:
global y;
global h;
y = (diff(log(p))-mean(diff(log(p))))*100;
h = zeros(size(y));
But I don't see any reason to keep them global so just don't make them global.
Also, you mentioned this code worked fine in Matlab. I was under the impression that you couldn't initialize global and persistent variables in Matlab which would make your code illegal in Matlab.