%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% This is practice script 2 for the Matlab tutorial%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

rand('state',37); % set arbitrary seed for uniform draws
randn('state',37); % set arbitrary seed for normal draws

tic;  % start stop watch

[fid]=fopen('c:\mlab_tut\mlab\logs\script2.txt','w');
if fid==-1;
   warning('File could not be opened');
   break
else;
   disp('File opened successfully');
end;

%load data from Excel:
load c:\mlab_tut\from_xl.txt;
data = from_xl; %rename dataset with a simpler name - optional
clear from_xl;% erase original (duplicate) data set - optional

% or from STATA:
load c:\mlab_tut\from_stata.raw;
data = from_stata;
clear from_stata;

% describe contents of data
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%1  runid           running observation id  
%2  orig_solid      original solicitor ID
%3  amount          contributions (2005 dollars)
%4  gender          solicitor Gender (1=male)
%5  race            solicitor Race (1=white)
%6  height          solicitor height (inches)
%7  weight          solicitor weight (lbs)
%8  beauty          beauty index 
%9  spunk           sum of personality scores
%10 gen_resp        observed gender of respondent (1=male)                                               
%11 raceresp        observed race of respondent(1=white)                                           
%12 avg_HHsize      average household size (Census block)
%13 inc000          Census tract median income (2005 dollars)                
%14 lottery         fundraising package included lottery component

% define dependent variable and matrix of explanantory variables
y = data(:,3); %":" as a dimension indicator means "all rows" or "all columns"; 
% here it means "all rows" since it describes the row dimension
X = data(:,4:end); % the second ":" means from - to; here from the 4th column to the last
% Used in this context "end" indicates the last row, column, or element of a numerical
% array
n=length(y);
k=size(X,2);

%Summary statistics:
my = mean(y);
stdy = std(y);
Vy = var(y);
Sy = sum(y);

out=[my stdy Vy Sy]; 
fprintf(fid,'Summary stats for y ("amount") \n'); 
fprintf(fid,'mean\t\tstd\t\tvariance\t\tsum\n'); 
fprintf(fid,'%6.3f\t%6.3f\t%6.3f\t%6.3f\n',out'); 
fprintf(fid,'\n'); 

mX=mean(X,1);%mean for each column; result will be 1 by k
% note: add a transpose at the end if you prefer to see this as a k by 1
stdX = std(X,1);%std for each column; result will be 1 by k
VX=var(X,1);% variance for each column; result will be 1 by k
covX=cov(X,1);% covariance across columns; result will be k by k
corrX=corr(X);% sample correlation across columns; result will be k by k
sX=sum(X,1);% sum for each column; result will be 1 by k


% call OLS function "ols1" and perform remaining regression computations in main
% script.
bols=ols1(y,X);

e=y-X*bols;
s2=e'*e/(n-k); 
Vb=s2*inv(X'*X); 
se=sqrt(diag(Vb));
t=bols./se; 

%or: call OLS function "ols2";
[bols,se,t,s2] = ols2(y,X);

out=[bols se t]; %combine bols, se, and t vectors into a single matrix
fprintf(fid,'Output table for betas \n'); 
fprintf(fid,'coeff\t\tstd\t\tt-value\n'); 
fprintf(fid,'%6.3f\t%6.3f\t%6.3f\n',out'); 
fprintf(fid,'\n');
fprintf(fid,'Squared regression error=\t%6.3f \n',s2); 
fprintf(fid,'\n');

save c:\mlab_tut\mlab\worksp\script2_stuff y X bols;

finish = toc; %this will show run time in seconds
fprintf(fid,'Time elapsed in seconds \n\n');
fprintf(fid,'%6.3f\n',finish);

st=fclose(fid);
if st==0;
   disp('File closed successfully');
else;
   warning('Problem with closing file');
end;