Event Study - Non Parametric Tests
Afficher commentaires plus anciens
I have run this code, but it does not yield the same result as in Excel?
%%%%%T4 Statistics Calculation
S_adj_T4 = S_adj_T3_2 ;
rep_S_adj_T4 = repmat(S_adj_T4, L1+L2, 1) ;
AR_all = [AR_est ; AR_event] ;
adj_stand_AR_T4 = AR_all./rep_S_adj_T4 ;
rank_T4 = tiedrank(adj_stand_AR_T4) ;
U_T4 = rank_T4/((L1+L2)+1) - 1/2 ;
for i = 1 : L1+L2 ;
sum_U_T4(i) = nansum(U_T4(i, :)) ;
end
for i = 1 : L1+L2 ;
nPos_T4(i) = nansum(U_T4(i,:)>0);
end ;
for i = 1 : L1+L2 ;
sqrt_T4(i) = 1/(sqrt(nPos_T4( : , i))) ;
end ;
q = nansum((sum_U_T4.*sqrt_T4).*(sum_U_T4.*sqrt_T4)) ;
S_k = sqrt(1/(L1+L2)*q) ;
for i = 1 : L1+L2 ;
T4stat(i) = (1/sqrt(N)*sum_U_T4( : , i))/S_k ;
end ;
T4stat = T4stat( : , L1+1:end) ;
for i = 1 : L1+L2 ;
T4CARstat(i) = (1/sqrt(N)*sum_U_T4( : , i))/(S_k*sqrt(L2)) ;
end ;
T4CARstat = nansum(T4CARstat( : , L1+1:end)) ;
3 commentaires
the cyclist
le 24 Août 2013
What's your question?
Daniel
le 24 Août 2013
the cyclist
le 24 Août 2013
See the replies here:
I would warn you, though, that I don't think you will get someone willing to go line-by-line through your code, and guess what is happening in Excel that might be different.
I think you might need to narrow your problem a little.
Réponses (1)
Daniel
le 25 Août 2013
0 votes
Catégories
En savoir plus sur Data Import from MATLAB dans Centre d'aide et File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!