After using mvregress, how can I find the rsquared value, t-values, p-values, F-statistic, and standard errors?
7 Ansichten (letzte 30 Tage)
Ältere Kommentare anzeigen
After using mvregress, how can I find the rsquared value, t-values, p-values, F-statistic, standard errors and residual sum of squares? I have run a fixed effect regression using panel data. I have gotten my values for the intercept and slope parametres, however, now I want more information about the model and I'm not sure how to go about it. Any help on this one?
0 Kommentare
Antworten (1)
Juan Zamora
am 8 Mär. 2015
Y = ChangePriceSqft;
X = [OneOverSqft, Baths, Distance];
% print scatter plot matrix againt sales
[H,AX,BigAx,P,PAx] = plotmatrix([Y,X],[Y,X]);
% regression model
X = [ones(n,1), X];
%[b, bint, r, rint, stats] = regress(Y, X);
[beta,Sigma,E,CovB,logL] = mvregress(X,Y);
Hypothesis = beta(1) + beta(2) * OneOverSqft + beta(3) * Baths + ...
beta(4) * Distance;
% r-squared
r2 = sum((Hypothesis - mean(Y)).^2) / sum((Y - mean(Y)).^2);
% number of explanatory variables
k = 3;
% Data from the Regression Model
% beta(1); % intercept
% beta(2); % 1/sqft
% beta(3); % baths
% beta(4); % distance
% model Standard Error
Se2 = sum((Y - beta(1) - beta(2) * OneOverSqft - beta(3) * Baths + ...
beta(4) * Distance).^2) / (n-k-1);
Se = sqrt(Se2);
% Partial Slopes - Standard Errors Bx
PartialStdErr = diag(sqrt(CovB)); % mvregress
%t-statistics (Beta / PartialStdErr)
tRatio = beta ./ PartialStdErr;
%p-values
pVals = 2*(1-tcdf(abs(tRatio),n-2));
% Summary Table of the Regression Model
RegressionSummary = [beta, PartialStdErr, tRatio, pVals];
1 Kommentar
Juan Zamora
am 11 Mär. 2015
You can find the entire code with the csv file at https://github.com/zamoradev/ML/tree/master/MRM_HomePrices
There are more samples at: https://github.com/zamoradev/ML
Siehe auch
Kategorien
Mehr zu Linear Regression finden Sie in Help Center und File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!