Regression constant

ruialmeida · Post by **ruialmeida** » Sun Oct 27, 2013 2:45 pm

Hello all,

The regression equation is
y= a + beta X, where a is a constant.

I would like to know where I can find this constant in the SmartPLS output.

How can I validate all usual, regression assumptions

Thanks in advance
Best regards

Hengkov · Post by **Hengkov** » Mon Oct 28, 2013 10:14 am

Hi,

Because SmartPLS using standardized data (Mean 0 Var 1), the constant value not available. PLS based on non-parametric/distribution free, so the classical linear assumption such normality cannot applied (just collinearity).

Best Regards,

ruialmeida · Post by **ruialmeida** » Tue Oct 29, 2013 6:24 pm

Thank you very much Hengkov,

if I use original data metric I'll get the constant?

Thank you in advance.
Best regards

Hengkov wrote:Hi,

Because SmartPLS using standardized data (Mean 0 Var 1), the constant value not available. PLS based on non-parametric/distribution free, so the classical linear assumption such normality cannot applied (just collinearity).

Best Regards,

Hengkov · Post by **Hengkov** » Tue Oct 29, 2013 8:20 pm

Hi,

Yes, acctually the unstandardized data will result constant value.
But, SmartPLS based on standardized.

Regards,