How to erase the errors from a Multivariable Polynomial Equation from Regression !!!

[gangwarss]

Hi. I am in need of some expert inputs.

I have run a regression on a dataset for a multivariable polynomial equation.

My p values and Significance F are fit but still the equation doesnt pass because of big error quotient.

My limited knowledge is that in excel you can only get uni variable polynomial equation or multi variable linear equation at maximum.

Can anyone help please?

Attaching my analysis. Thanks in Advance.

[gangwarss]

Please take a look at the file. The issue will be more clear to understand.

[Glenn Kennedy]

I did... and it isn't. Please explain what is going on.

[gangwarss]

Hi Glenn,

Thanks for responding.

So in the file my primary data is in (Row B to F). Y is the dependent Variable and X1,2,3,4 are independent.

Row H to L is the polynomial conversion of the primary data. So the equation is , Y = aX1^5 + bX2^(1/5) + cX3 + dX4 + m.

The Regression on Q14 is the regression outcome for Dataset Row H to L.

Can you please explain why is my Equation not getting passed? What am I ignoring ? My p values are good. My significance F is low. where is the issue?

Regards,

[gangwarss]

Hope not it is clear than before?

[MrShorty]

I don't see anywhere in the spreadsheet where you are calculating an "error quotient", so I'm not sure what you are referring to when you refer to the error quotient. I notice that the X4 values in column F are very different from the X4 values in column L, but I do not know if that is intended or not. Other than that, I see nothing wrong with the programming aspects of this. If the mismatch between column F and column L is not the problem, then I suspect the problem lies in something about the data and/or model that we do not know on this side of the internet.