I have performed regression analysis in excel 2007. Can any one help me to interpret the result? What to do, when the coefficient and t stat are negative? Please examine the attached ms word document containing the output of regression function obtained in excel 2007. Thanks in advance.
The t-value is the value of the coefficient divided by its standard error. Therefor the t-value always has the same sign as the coefficient.
A negative coefficient means that when the value of the corresponding predictor increases, then the expected response decreases.
In your analysis, the coefficients for X1 and X3 are not very precisely estimated. This means, the data you used to fit the model is sighltly more likely for positive values of X1 and for negative values of X3. The coefficient for X2 is the most clear one. The data is most likely for positive values of this parameter, negative values are really unreasonable.
What cannot be seen from your tables is if the model is sensible at all and if the distribution of the residuals is supporting the above interpretations. H suggest appropriate residual plots to see if the residual distribution is symmetric and shows no obvious patterns depending on the predictors or the response.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression