how to do vector

Hi,

I have code to run but firstly my data should be balanced panel which I did. however, the second condition is that each variable a N*T vector which I do not know how to do. Could you please advise me how to do this task with thanks

Best regards

[Birdman]

Can you share your code?

[Maria hassan]

Hi

Here is the code:my question is how I enter the series which contains around 10 series since I cannot name all the series same name.

function[x]=pd2(series,N,T)

psi=reshape(series,T,N);

z=psi(2:T,:)-psi(1:T-1,:);

x=reshape(z(2:T-1,:)-z(1:T-2,:),N*(T-2),1);

thanks

Answer 1

[John BG]

Hi Maria

this is John BG <mailto:[e-mail] [e-mail]>

the problem is the line

psi=reshape(series,T,N);

inside the function pd2.

reshape only works if the input matrix has exactly N*T elements.

To solve the uncertainty, you can do something like the following:

clear all;close all;clc

N=4;T=5;
series=randi([0 10],1,randi(N*T*2,1))
numel(series)
psi=0;

the previous is just to test it works.

Use something like these 2 ifs: if length of series shorter than N*T, then pad zeros or a value of your choice until you have N*T elements, and then use reshape.

If series has more elements than N*T you have make a decision, a possible policy could be take the N*T initial elements of the input series.

if numel(series)<=N*T   % if series  length shorter that N*T, add zeros up to N*T
    psi=[series zeros(1,N*T-numel(series))]
    psi=reshape(psi,T,N);
end

if numel(series)>N*T     % if series length larger than N*T, 
    psi=series([1:N*T])
    psi=reshape(psi,T,N);
end

If you find this answer useful would you please be so kind to consider marking my answer as Accepted Answer?

To any other reader, if you find this answer useful please consider clicking on the thumbs-up vote link

thanks in advance for time and attention

John BG

<mailto:[e-mail] [e-mail]>

[Maria hassan]

Hi John,

Many thanks for your reply.

my data is balanced data. I have 18*16=288. but How can I run the code I could not get it. so when I import the data, I import them as matrix and name the matrix as series so it work? I have run it and the results is column with 252 elements but elements from 238 till 252 are NaN. any idea why s that?

Regards Maria

[John BG]

The NaN elements are not a problem for reshape, example

A=[1 NaN 3 7 -1 ; 0 NaN 5 8 11]
A =
     1   NaN     3     7    -1
     0   NaN     5     8    11
 reshape(A,1,10)
 =
     1     0   NaN   NaN     3     5     7     8    -1    11

you see? reshape knows it has to take then into account as if they were values.

As you are pointing out, the input matrix sometimes is not 288, but only 252.

This is why you have to fill up to 288 with more NaN elements, or zeros, or an element you choose convenient.

Add this

if numel(series)<=N*T  
    psi=[series zeros(1,N*T-numel(series))]
    psi=reshape(psi,T,N);
end

to solve when the input is shorter than expected, to allow reshape to work correctly.

Use

if numel(series)>N*T     
    psi=series([1:N*T])
    psi=reshape(psi,T,N);
end

to ignore the excess when the input is larger than the expected size.

Would it be possible for you to show the input data itself?