selecting to do operation on every fourth row

this quetion might sound confusing

i have the 840 rows of the following is a sample

          f(:,1) f(:,2) f(:,3)
  air     0.693  0.329  0.013  
  train   0.109  0.203  0.168   
  bus     0.356  0.153  0.250    
  car     0.010  0.076  0.193  

this the previous function i had which was easy

fun = @(x)[(((1+x(4))*(x(1)*f(:,1) + x(2)*f(:,2) + x(3)*f(:,3))) + (x(4)*((f(:,1).^x(1)).*(f(:,2).^x(2)).*(f(:,3).^x(3)))) + ep - en - u);

the question i like to code the following for all the my data. i want to do the following opation on every 4 rows grouped togther

to explain further

x(1)*((column1,row1) -  (column1,row2))+ x(2)*((column2,row1)-(column2,row2))+ x(3)*((column3,row1)-(column3,row2))
x(1)*((column1,row1) -  (column1,row3))+ x(2)*((column2,row1)-(column2,row3))+ x(3)*((column3,row1)-(column3,row3))
x(1)*((column1,row1) -  (column1,row4))+ x(2)*((column2,row1)-(column2,row4))+ x(3)*((column3,row1)-(column3,row4))

i am basically taking the first row minus the other rows in each column

this will repeat for the set of 4 untill i get to the end of 840 enteries

i hope someone can help

Answer 1

[Samatha Aleti]

Hi,

As per my understanding you want to consider every 4 rows as 1 set , apply your formula and repeat this till the last row. You can use a vector as “index” with increments of 4 and define the logic (for formula) using this “index” accordingly. Here is a sample code:

f = randi(5,[840 3]); % Consider random array of size 840 x 3 
y = zeros(840,1); % Array to store output 
i= 1: 4: 840; % Index  
y(i) = (f(i,1) - f(i+1,1)) + (f(i,2) - f(i+1,2)) + (f(i,3) - f(i+1,3)); 
y(i+1) = (f(i,1) - f(i+2,1)) + (f(i,2) - f(i+2,2)) + (f(i,3) - f(i+2,3)); 
y(i+2) = (f(i,1) - f(i+3,1)) + (f(i,2) - f(i+3,2)) + (f(i,3) - f(i+3,3)); 
y(i+3) = NaN; % Ignore

[Abdulaziz Altowijri]

i think this is similar to what i am looking for is it easy to adapt this to my code

%% Optimization - Minimizing Sum of Squared Errors

% Minimum SSE = sum(ep)^2 + sum(en)^2

% Clearing Command Window

clc

% Clearing Variables from Workspace

clearvars

close all

% Reading Data from Excel File & Sotring in variable "f"

f = xlsread('data.xlsx');

% Finding Size of Data

[m, n] = size(f);

% % Taking Input from user for values of U

% fprintf('\n\tEnter Value of U\t:\t')

% u = input('');

% Defining Lower Limits of [beta alpha w(1) w(2) w(3)]

% beta > 0

% alpha (Unrestricted in sign)

% w(i) >= 0

lb = [0.01 -inf 0 0 0];

% Defining Upper Limits of [beta alpha w(1) w(2) w(3)]

% beta < infinity

% alpha (Unrestricted in sign)

% w(i) <= 1

ub = [inf inf 1 1 1];

%Initial values of ep & en

ep=abs(0.1*rand(840,1));

en= abs(0.1*rand(840,1));

% Guessing Initial Values of [beta alpha w(1) w(2) w(3)]

x0 = [1e-1 0 0.6 0.2 0.2];

u=0.25*ones(840,1);

% Defining Function for optimization

%% here is where i need to use the index part here i am using to work with the the lsqnonlin function

fun = @(x)[x(3)*f(:,1).^x(1) + x(4)*f(:,2).^x(1) + x(5)*f(:,3).^x(1)+ep-en - x(2)*u;

x(3) + x(4) + x(5)-1];

[x,SSE] = lsqnonlin(fun,x0,lb,ub,options);