Not enough input arguments.

Hi, i am having issue to run a VAR model were the following error is flagged:

>> VARmodel

Not enough input arguments.

Error in VARmodel (line 33)

[nobs, nvar] = size(ENDO);

Should I have to quote nobs and nvar?

[Image Analyst]

You forgot to attach VARmodel(). It looks like it expects an input. What does the function definition say

function result = VARmodel(someInputVariable)

In your code, what is someInputVariable called and how did you call VARmodel()? You didn't just click the green run triangle did you? That does not pass in anything for the required input variable.

[economics student]

Yes I did :D apologies. Below you have the coding but i cannot see the function as a result except for the companion VAR that is only a partial result. Maybe i should try to go for the example:

Perform vector autogressive (VAR) estimation with OLS

%========================================================================
% [VAR, VARopt] = VARmodel(ENDO,nlag,const,EXOG,nlag_ex)
% -----------------------------------------------------------------------
% INPUT
%	- ENDO: an (nobs x nvar) matrix of y-vectors
%	- nlag: lag length
% -----------------------------------------------------------------------
% OPTIONAL INPUT
%	- const: 0 no constant; 1 constant; 2 constant and trend; 3 constant 
%       and trend^2 [dflt = 0]
%	- EXOG: optional matrix of variables (nobs x nvar_ex)
%	- nlag_ex: number of lags for exogeonus variables [dflt = 0]
% -----------------------------------------------------------------------
% OUTPUT
%   - VAR: structure including VAR estimation results
%   - VARopt: structure including VAR options (see VARoption)
% -----------------------------------------------------------------------
% EXAMPLE
%   - See VARToolbox_Code.m in "../Primer/"
%% Check inputs
% -----------------------------------------------------------------------
[nobs, nvar] = size(ENDO);

% Create VARopt and update it
VARopt = VARoption;
VAR.ENDO = ENDO;
VAR.nlag = nlag;

% Check if ther are constant, trend, both, or none
if ~exist('const','var')
    const = 1;
end
VAR.const = const;

% Check if there are exogenous variables 
if exist('EXOG','var')
    [nobs2, nvar_ex] = size(EXOG);
    % Check that ENDO and EXOG are conformable
    if (nobs2 ~= nobs)
        error('var: nobs in EXOG-matrix not the same as y-matrix');
    end
    clear nobs2
    % Check if there is lag order of EXOG, otherwise set it to 0
    if ~exist('nlag_ex','var')
        nlag_ex = 0;
    end
    VAR.EXOG = EXOG;
else
    nvar_ex = 0;
    nlag_ex = 0;
    VAR.EXOG = [];
end


%% Save some parameters and create data matrices
% -----------------------------------------------------------------------
    nobse         = nobs - max(nlag,nlag_ex);
    VAR.nobs      = nobse;
    VAR.nvar      = nvar;
    VAR.nvar_ex   = nvar_ex;    
    VAR.nlag      = nlag;
    VAR.nlag_ex   = nlag_ex;
    ncoeff        = nvar*nlag; 
    VAR.ncoeff    = ncoeff;
    ncoeff_ex     = nvar_ex*(nlag_ex+1);
    ntotcoeff     = ncoeff + ncoeff_ex + const;
    VAR.ntotcoeff = ntotcoeff;
    VAR.const     = const;

% Create independent vector and lagged dependent matrix
[Y, X] = VARmakexy(ENDO,nlag,const);

% Create (lagged) exogenous matrix
if nvar_ex>0
    X_EX  = VARmakelags(EXOG,nlag_ex);
    if nlag == nlag_ex
        X = [X X_EX];
    elseif nlag > nlag_ex
        diff = nlag - nlag_ex;
        X_EX = X_EX(diff+1:end,:);
        X = [X X_EX];
    elseif nlag < nlag_ex
        diff = nlag_ex - nlag;
        Y = Y(diff+1:end,:);
        X = [X(diff+1:end,:) X_EX];
    end
end
%% OLS estimation equation by equation
% -----------------------------------------------------------------------
for j=1:nvar
    Yvec = Y(:,j);
    OLSout = OLSmodel(Yvec,X,0);
    aux = ['eq' num2str(j)];
    eval( ['VAR.' aux '.beta  = OLSout.beta;'] );  % bhats
    eval( ['VAR.' aux '.tstat = OLSout.tstat;'] ); % t-stats
    eval( ['VAR.' aux '.bstd  = OLSout.bstd;'] );  % beta std error
    % compute t-probs
    tstat = zeros(ncoeff,1);
    tstat = OLSout.tstat;
    tout = tdis_prb(tstat,nobse-ncoeff);
    eval( ['VAR.' aux '.tprob = tout;'] );        % t-probs
    eval( ['VAR.' aux '.resid = OLSout.resid;'] );% resids 
    eval( ['VAR.' aux '.yhat  = OLSout.yhat;'] ); % yhats
    eval( ['VAR.' aux '.y     = Yvec;'] );        % actual y
    eval( ['VAR.' aux '.rsqr  = OLSout.rsqr;'] ); % r-squared
    eval( ['VAR.' aux '.rbar  = OLSout.rbar;'] ); % r-adjusted
    eval( ['VAR.' aux '.sige  = OLSout.sige;'] ); % standard error
    eval( ['VAR.' aux '.dw    = OLSout.dw;'] );   % DW
end 


%% Compute the matrix of coefficients & VCV
% -----------------------------------------------------------------------
Ft = (X'*X)\(X'*Y);
VAR.Ft = Ft;
F = Ft';
VAR.F = Ft';
SIGMA = (1/(nobse-ntotcoeff))*(Y-X*Ft)'*(Y-X*Ft); % adjusted for # of estimated coeff per equation
VAR.sigma = SIGMA;
VAR.resid = Y - X*Ft;
VAR.X = X;
VAR.Y = Y;
if nvar_ex > 0
    VAR.X_EX = X_EX;
end


%% Companion matrix of F and max eigenvalue
% -----------------------------------------------------------------------
Fcomp = [F(:,1+const:nvar*nlag+const); eye(nvar*(nlag-1)) zeros(nvar*(nlag-1),nvar)];
VAR.Fcomp = Fcomp;
VAR.maxEig = max(abs(eig(Fcomp)));


%% Initialize other results
% -----------------------------------------------------------------------
VAR.B   = [];   % structural impact matrix (need identification: see VARir/VARfevd)
VAR.b   = [];   % first columns of structural impact matrix (need identification: see VARir/VARfevd)
VAR.PSI = [];   % Wold multipliers (computed only with VARir/VARfevd)
VAR.Fp  = [];   % Recursive F by lag (useful to compute MA representation)
VAR.IV  = [];   % External instruments for identification

[Image Analyst]

OK, the function definition line is in a comment:

% [VAR, VARopt] = VARmodel(ENDO,nlag,const,EXOG,nlag_ex)

so it's not a function, it's now a script. So you need to define ENDO,nlag,const,EXOG, and nlag_ex in advance. What values did you assign to them? Show that code.

[economics student]

Ok. thanks.

I did not define them as i thought that these would adjust depending on the data imported.

[Stephen23]

Do NOT use eval for trivial field access. Instead of this ugly, slow, inefficient, complex code:

eval( ['VAR.' aux '.beta  = OLSout.beta;'] );

use this simpler, neater, more efficient, and much easier to debug code:

VAR.(aux).beta  = OLSout.beta;

https://www.mathworks.com/help/matlab/matlab_prog/generate-field-names-from-variables.html