Exponential of a matrix

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,

There are several ways to compute the exponential of a matrix. One of
these ways consists in using the spectral decomposition of the matrix,
that the HP50g can do.

As a result, why is EXP(A), where A is a matrix, not implemented? This
would be very useful, especially for systems theory.
- -- 
Merciadri Luca
See http://www.student.montefiore.ulg.ac.be/~merciadri/
- -- 

Procrastination is the thief of time.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.8 <http://mailcrypt.sourceforge.net/>

iEYEARECAAYFAk2HQB0ACgkQM0LLzLt8MhyVhwCglF+9ofQFo2GWGaX70QuSYawZ
K1MAnR+nKqBh/pGiWr62zOEz7zyJJkOy
=miqb
-----END PGP SIGNATURE-----
0
Merciadri
3/21/2011 12:10:05 PM
comp.sys.hp48 9721 articles. 11 followers. fbarbaise (60) is leader. Post Follow

4 Replies
738 Views

Similar Articles

[PageSpeed] 44
On 21 Mar, 13:10, Merciadri Luca <Luca.Mercia...@student.ulg.ac.be>
wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Hi,
>
> There are several ways to compute the exponential of a matrix. One of
> these ways consists in using the spectral decomposition of the matrix,
> that the HP50g can do.
>
> As a result, why is EXP(A), where A is a matrix, not implemented? This
> would be very useful, especially for systems theory.
> - --
> Merciadri Luca
> Seehttp://www.student.montefiore.ulg.ac.be/~merciadri/
> - --
>
> Procrastination is the thief of time.
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.9 (GNU/Linux)
> Comment: Processed by Mailcrypt 3.5.8 <http://mailcrypt.sourceforge.net/>
>
> iEYEARECAAYFAk2HQB0ACgkQM0LLzLt8MhyVhwCglF+9ofQFo2GWGaX70QuSYawZ
> K1MAnR+nKqBh/pGiWr62zOEz7zyJJkOy
> =miqb
> -----END PGP SIGNATURE-----

You can use DIAGMAP command, see HP 50g AUR http://www.hpcalc.org/details.php?id=7141
at page 3-58:

ex. [[ 1 1 ] [ 0 2 ]] 'EXP(X)' DIAGMAP --> [[ 'EXP(1)' '-
EXP(1)+EXP(2)' ][ 0 'EXP(2)' ]]

Simone Cerica
0
supergems
3/21/2011 4:37:08 PM
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

supergems <simone.cerica@gmail.com> writes:

> On 21 Mar, 13:10, Merciadri Luca <Luca.Mercia...@student.ulg.ac.be>
> wrote:
>> -----BEGIN PGP SIGNED MESSAGE-----
>> Hash: SHA1
>>
>> Hi,
>>
>> There are several ways to compute the exponential of a matrix. One of
>> these ways consists in using the spectral decomposition of the matrix,
>> that the HP50g can do.
>>
>> As a result, why is EXP(A), where A is a matrix, not implemented? This
>> would be very useful, especially for systems theory.
>> - --
>> Merciadri Luca
>> Seehttp://www.student.montefiore.ulg.ac.be/~merciadri/
>> - --
>>
>> Procrastination is the thief of time.
>> -----BEGIN PGP SIGNATURE-----
>> Version: GnuPG v1.4.9 (GNU/Linux)
>> Comment: Processed by Mailcrypt 3.5.8 <http://mailcrypt.sourceforge.net/>
>>
>> iEYEARECAAYFAk2HQB0ACgkQM0LLzLt8MhyVhwCglF+9ofQFo2GWGaX70QuSYawZ
>> K1MAnR+nKqBh/pGiWr62zOEz7zyJJkOy
>> =miqb
>> -----END PGP SIGNATURE-----
>
> You can use DIAGMAP command, see HP 50g AUR http://www.hpcalc.org/details.php?id=7141
> at page 3-58:
>
> ex. [[ 1 1 ] [ 0 2 ]] 'EXP(X)' DIAGMAP --> [[ 'EXP(1)' '-
> EXP(1)+EXP(2)' ][ 0 'EXP(2)' ]]
Thanks!

- -- 
Merciadri Luca
See http://www.student.montefiore.ulg.ac.be/~merciadri/
- -- 

The nail that sticks up will be hammered down.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.8 <http://mailcrypt.sourceforge.net/>

iEYEARECAAYFAk2HgxUACgkQM0LLzLt8Mhz5FQCfbX/rsQoqV2GWEdhJr5D+lwIO
/CAAn1wCN5hyIgTGTgj07WBWGa19L7X7
=ju2t
-----END PGP SIGNATURE-----
0
Merciadri
3/21/2011 4:55:50 PM
When the matrix is not diagonalizable you can use this code:

\<< DUP SIZE OBJ\-> DROP2 XQ IDN
RCLVX * SWAP - INV ILAP EXPAND
\>> 'MATEXP' STO
 
[[ 3 0 0 ][ 0 2 1 ][ 0 0 2 ]] MATEXP --> [[ 'EXP(3*X)' 0 0 ][ 0 'EXP(2*X)' 'X*EXP(2*X)' ][ 0 0 'EXP(2*X)' ]]
[[ 1 -1 ][ 0 2 ]] MATEXP --> [[ 'EXP(X)' '-(EXP(2*X)-EXP(X))' ][ 0 'EXP(2*X)' ]]
0
supergems
1/7/2015 10:24:18 PM
For numerical evaluation you could try 
matexp at
http://www.hpcalc.org/hp49/math/numeric/utilgcc.zip 

quite fast, written in hpgcc2, runs on the hp50g


best regards
Gjermund
0
gjermund
2/16/2015 11:03:39 AM
Reply:
Similar Artilces:

Transition Matrix (Markov)
Hi, I'm using homogenous Markov Chain model to predict the future employees' movement in the organization. I'm new in using SAS and I have few questions. Q-1I have Table which has -Emplid, Effective date, Action. I would like a table that returns a single record for each Emplid with their most recent effective date for the Action. Emplid Effective date Action 0001 2007-03-02 Hire 0001 2007-05-02 Ter 0003 2007-11-12 Hire 0005 2007-04-12 Rehire 0005 2007-03-24 Term I...

DCT of an 8x8 matrix?
I'm trying to apply the DCT as described in [1], but it isn't working well.* Please take a look at my Maple worksheet in [2] (it is an HTML export). I would greatly appreciate a pointer as to what I could be doing wrong. As far as I can tell I am implementing the FDCT correctly. Thank you, James Sumners [1] -- http://www.ijg.org/files/wallace.ps.gz [2] -- http://student.claytonstate.net/~jsumners/DCT/index.html * Actually, I'm not shifting the unsigned integers to signed. But I still shouldn't be getting the result I am getting I don't believe. James Sumners wrote...

Isolating matrix data
Hello users, What is the best way to extract data from a large matrix (A) if you know the row and column numbers (rn,cn) of the data you want to extract? For my data, size(A)=[512 512] and length(rn)=20,000 I've tried using : for i=1:length(rn); data_vector(i)=A(rn(i),cn(i)); end but it takes too long, and using: data_vector=A(rn,cn) doesn't work either. Any suggestions? Thanks, Matt K. Matt K. wrote: > > > Hello users, > > What is the best way to extract data from a large matrix (A) if you > know the row and column numbers (rn,cn) of the data you want to ...

Matrix #42
Hi, I have a matrix eg. [1 1 1; 1 2 1; 1 1 1] i want to create a formula that looks up the highest value in the matrix (ie.2) eg. V = 0.5 * (highest value in the matrix) Can anyone help? Thank you in advance. On 9/18/2013 7:35 AM, Sarwar wrote: > Hi, > > I have a matrix eg. [1 1 1; 1 2 1; 1 1 1] > > i want to create a formula that looks up the highest value in the matrix (ie.2) > eg. V = 0.5 * (highest value in the matrix) > > Can anyone help? > Thank you in advance. > V = 0.5*max(A(:)) ...

transforming exponential of sums into product of exponentials
Dear All, I can not understand why Mathmatica refuses to apply rules which transform exponenials of sums into product of exponentails. For instance, f[x+y] /. f[a_ + b_]-> f[a] f[b] returns f[x] f[y], whereas Exp[x+y] /. Exp[a_+b_]->Exp[a]Exp[b] still returns Exp[x+y] Any help greatly appreciated Hi, it does it because it also evaluate 2*2 to 4. You must use a Hold[] to prevent the simplifications Exp[x + y] /. Exp[a_ + b_] :> Hold[Exp[a]Exp[b]] Regards Jens Karim Mehadhebi wrote: > > Dear All, > > I can not understand why M...

help loading data as a matrix
I need to convert data from a text file into a matrix. The way the data is stored in our database is (RowName, ColumnName, Value). Sally, Joe, 3 Sally, Bill,4 Joe, Bill,8 Sally, Sally,2 Joe, Joe, 5 Bill, Bill,10 This data needs to be transformed so I get a 3x3 matrix: [2 3 4; 0 5 8; 0 0 10] Hopefully preserving or returning the order of columns and rows. Sally, Joe, Bill Sally Joe Bill I�ve been trying textread and textscan and don�t think either of these will do the trick. I was hoping to use some sort of index to assign the first value as the Row, the second value the Column an...

algorithm with exponential deviates
dear all i am trying to find the number of preferences one has to state if 100 candidates apply for 100 different jobs. This means we want to match candidates to jobs. The candidates have exponential preferences, ie many candidates want the jobs with low indexes. i is the candidate number job is the job assigned x1-x100 are the preferences My algorihtm unfortunately exceeds 100 and runs 'forever'. I cannot spot the error. Dirk options mprint mlogic; %let jobs=100; data ran; do i=1 to &jobs; x1=round(&jobs/4*ranexp(1),1); output; end; run; pr...

sas tech support tip: Character values translated to Exponential
fwd by a Concerned Reader from our city user group: Greater Atlanta SAS User Group GASUG http://www.gasug.org/ > I just remembered that I was telling you > of a strange problem I had - and the accompanying strange solution! > I was printing a list of DASD devices > and the device number is in HEX. > In the program I stored it as a 4-character alphanumeric variable. > Still, certain values, > when sent through tagsets.ExcelXP on z/OS, then emailed, > would show up in exponential notation. > Here is the suggestion I got from SAS Tech Support > - and it works! Jus...

Exponential of very small numbers
Hello, first post ;) i need to integrate the planck distribution on IDL, but i have a problem with the exponential. The error is: EXP(1.6671699e-013)=1 What can i do to avoid this problem? Even if it's an exponent very close to zero, it's important for my code. Thanks! For x very close to zero I would use the approximation exp(x) = x or you could display all the digits (and use double precision) IDL> print,EXP(1.6671699d-013)-1 1.6675550e-13 IDL> print,EXP(1.6671699d-013),f='(f24.18)' 1.000000000000166755 --Wayne On Wednesday, May 8,...

Theory of exponential (power set) turing machines --- why not do it ?
Hi, Whatever these quantum dots are ----- do they compute the halting problem of a turing machine. Now, irrespective of that ----- why don't we have the theory of exponential turing machines (power set turing machines) the power set of power set of turing machines and so on. I mean, the halting problem for a turing machine can be solved by a turing machine which is the power set of a turing machines ---------------- this is clearly known as an exponential turing machine to an exponential problem. Now, one person says, it must be representable in finite space ------- well how about infini...

Re: A complicated matrix manipulation question
The notation is quite hard to read. Try breaking the problem down into the computational sequence; that is, what nedds to be calculated first, second, ... Then try to implement in that order, again one step at a time (you can always optimize later). On Mon, 2 Aug 2004 08:18:33 -0700, Feixia Cheng <feixiacheng@YAHOO.COM> wrote: >The data I have is of the following structure: > >id W1 W2 >1 10 12 >1 11 14 >1 6 7 >2 6 9 >2 11 14 >3 15 14 >3 13 12 >........ > >Now call vector W_{ij}=(W1_{ij}, W2_{ij}), I need to first ...

Contour matrix objects help, Please
Hello all, The output of Matlab's contour or contourm function is a plot and also a 2N tall matrix of contour coordinates, where N is the number of contour lines. I would like to figure out how to use these coordinates. I ran contour on a simple matrix: c = contour( ... [ 0 0 0 1 ;... 0 0 1 1 ;... 0 0 1 1 ;... 0 0 0 1 ] ) The output is: c = 0.5000 3.5000 3.0000 2.5000 2.5000 3.0000 3.5000 6.0000 4.0000 3.5000 3.0000 2.0000 1.5000 1.0000 What do these numbers mean? What format ...

How to "search" in a matrix?
I have this matrix: matr = {{a, b, 1}, {c, d, 2}, {e, f, 3}} I'm trying to do a "search" in a matrix. The "real" matrix will contain strings. How to "search" for a given value in a matrix and return all lines that have it? 'Select' seems to handle only lists - and this would require a Flatten, probably making my code too complicated and maybe slow (the real matrix will contain something like 100 elements - weather data). Thanks. Renan wrote: > I have this matrix: > > matr = {{a, b, 1}, {c, d, 2}, {e, f, 3}} > > ...

mex matrix of other types than mxReal #2
There is #include "matrix.h" mxArray *mxCreateDoubleMatrix(mwSize m, mwSize n, mxComplexity ComplexFlag); where ComplexFlag can be either mxREAL or mxCOMPLEX. How can I create matrixes to be of type 'single'? Thank you very much. "Mykola " <remove.this.nparfenukoff@mail.ru> wrote in message <g0bu9g$n3g$1@fred.mathworks.com>... > There is > #include "matrix.h" > mxArray *mxCreateDoubleMatrix(mwSize m, mwSize n, > mxComplexity ComplexFlag); > > where > > ComplexFlag can be either mxREAL or mxC...

Get 'paired indices' from 'matrix indices' (to create a sparse).
I think my logic is failing me. Curently i have: >K(dof,dof) = K(dof, dof) + k; Where K is a Sparse, and k is a length(dof) x length(dof) matrix. To speed things up, i want to 'vectorize' the indices so I can create the sparse properly, but for the life of me, I can't figure out how to do it smartly. ...

Using only upper triangular part of matrix to divide another matrix
Right now I am attempting to optimize my code as much as possible. I'm trying to optimize the code that does a./T. Is there any way to get the same results while only using the upper triangular portion of T or will that not help get speed gains? Also in order to deal with the NaN's on the resulting diagonal, after I divide I just set everything on the diagonal of the new matrix to 0. An example of what a and T look like: a = 0 -1 -2 -3 -4 1 0 -1 -2 -3 2 1 0 -1 -2 3 2 1 0 -1 4 3 2 1 0 T = ...

converting double number to exponential number
Is there any method in java to convert double number to its exponential equivalent with specific number of digits after decimal?? sweetone wrote: > Is there any method in java to convert double number to its exponential > equivalent with specific number of digits after decimal?? <http://java.sun.com/j2se/1.5.0/docs/api/java/text/DecimalFormat.html> HTH Andrew T. ...

histogram of vector frequency in matrix?
Hi. I'm trying to create a histogram to count the frequency of vectors (rows) in a matrix (MxN). For example, say I have a matrix that looks like the following: 1 2 3 4 5 6 7 8 9 1 2 3 1 2 3 I want to produce a histogram that displays that the vector [1 2 3] appears 3 times, and all others only once. This seems like it should be fairly simple, but I'm new to Matlab and can't figure it out. Any help would be appreciated. Thanks. "Jacob " <mymagdalina@gmail.com> wrote in message <gcsjp7$h9u$1@fred.mathworks.com>... > Hi. I'm trying to create a histogra...

how to generate a special matrix?
Hi: How can I generate the following binary matrix including all the binary codewords from 0 to N: 0 0 0 0 0 0 0 1 0 0 1 0 0 0 1 1 0 1 0 0 0 1 0 1 .... .... -- **************************************************** Gui Xie Graduate School of Information Science Japan Advanced Institute of Science and Technology Tatsunokuchi, Ishikawa, 923-1292, Japan Email: g-xie@jaist.ac.jp TEL: +81-090-62720598 FAX: +81-0761-51-1149 Web: http://www.jaist.ac.jp/~g-xie Gui Xie wrote: > Hi: > How can I generate the following binary matrix including all the binary > codewords from 0 to N: &g...

Exponential fitting
Hello, I'm trying to fit experimental data to an function of the form E(r) = D*(1-exp(a*(r-rm))^2 (i's the Morse potential from physics..). So far, I've written two scripts.. _____________________ x = load('xdata.dat'); y = load('ydata.dat'); rm = 1.3420; E0 = -455.7607; guess = rand(2,1); guess = guess'; coeff = lsqnonlin(@gafit,guess,rm,E0,x,y); x_fit = [0.5:0.05:3]; a=coeff(1); b=coeff(2) y_fit = a*(1-exp(-b*(x-rm))).^2 + E0; plot(x,y,'*',x_fit,y_fit,'r') __________________ and the 2nd one.. ____________________ fun...

What does this matrix code do?
Hello. I'm struggling to work out what bit of matrix manipulation the following piece of code is doing to the 3x3 matrix given: #include <stdio.h> int main() { int check_mat[3][3] = { { 1 , 2 , 3 } , { 4 , 5 , 6 } , { 7 , 8 , 9 } }; int im,jm; /* Print starting values to screen */ for ( im = 0 ; im < 3 ; im++ ) { for( jm = 0 ; jm < 3 ; jm++ ) { printf("%d\t" , check_mat[im][jm] ); } printf("\n"); } /* Perform some kind of matrix manipulation */ for( im = 0 ; im <= 2 ; im++ ) { for( jm = 0 ; jm &...

how to refine the u matrix in ss function?
sys = ss(A,B,C,D); can I rewrite the u matrix which premultiplied by matrix B? ...

how to make dynamic matrix(urgent)
I have ma matrix a= [1 2 3 4 5 6 6 7 3 2 1 4 6 7 6 7] if I enter the value 4 then program should should generate two matrix having element from above matrix. like b=a(:,even column) and c =a(:,odd column) if I enter if I enter the value 2 then program should should generate 4 matrix having element from above matrix. like b1=b(:,even column) and b2 =b(:,odd column) c1=c(:,even column) and c2 =c(:,odd column)....where b and c are derived form method first......how can I apply this method. Nilesh Joshi wrote: &g...

Matrix by cube multiplication, way to avoid for?
Hi everyone, I've been working around this problem for some days and I cannot go through this (hopefully last) point. For the first part of my problem I coudl simplify the calculations to a standard 2-d matrix multiplication: Result (m x n)= AuxMat (m x k) * dataMatrix (k x n) AuxMat has only 0s and 1s (for the moment, it might have real quotients later). I defined the matrix as sparse, as there is only one value per column. But, I also have to do a very similar calculation, but the dataMatrix has an extra dimension q, and I need this: Result (m x n x q)= somefunctionof (dataMatrix (...

Making Complex Matrix From Phase and Amplitude Matrix
I understand that, given two real matrices Real and Imaginary, you can make a complex matrix using ComplexMatrix=complex(Real, Imaginary); What if you have two matrices, Amplitude and Phase where Amplitude is a matrix giving the required amplitude values and Phase is a matrix giving the required unwrapped phases. Can you make a complex matrix from Amplitude and Phase? Phase is obtained from another complex matrix using Phase=unwrap(angle(OtherComplexMatrix)); Many thanks in advance, Peter. PeterOut <MajorSetback@excite.com> wrote in message <f118dd75-9ce9-4799- ad...