Interpreting & checking multiple regression code & output

Question

Jen 2019-11-22

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/492566-interpreting-checking-multiple-regression-code-output

回答： Pravin Jagtap 2019-11-25

ProjectMatrix.xlsx

I have the attached matrix and need to run a multiple regression. Columns 6 & 7 represent the coefficient of variation for conditions 1 & 2 respectively (condition 1 = no cognitive loading, condition 2 = cognitive loading), and my regression DV will be value differences between the two (i.e. column 7 - column 6). My two predictors will be age (column 2) and the percentage of cognitive loading trials (column 8). I have used the following code to run the regression but I'm not sure how to check if my code & output are correct.

Matrix attached

Code:

-----------------------------------

y = (matrix(:,7) - matrix(:,6));
ones = ones(200);
age = matrix(:,2);
loadpercentage = matrix(:,8);
X = [ones age loadpercentage];
[B,BINT,R,RINT,STATS] = regress(y, X)

-----------------------------------

Output:

STATS =

0.0026 0.2526 0.7770 0.0005

-----------------------------------

If I'm interpreting this correctly, the stats relate to the following:

Rsq value = 0.0026

F value = 0.2526

P value = 0.7770

estimated error variance = 0.0005

However, I'm not entirely sure how to interperet the figures. For example, my (limited) knowledge of regression tells me Rsq value relates to how much variance is explained by the model, but is 0.0026% correct? I think also that the F and P values are telling me the overall regression is not sinificant (at 0.7770), but again I have no idea if this is because I've coded the analysis incorrectly or if the regression in this case really is not significant.

For a little background info this assignment has been set during a teaching strike so I'm feeling my way through the stats in the dark