R^2 typically increases when you add variables to the model, so it's not a good measure of model prediction power in general, but particularly in this example. A better way to evaluate the value of a quadratic model is to look at the residuals, which is also possible using Excel. Always plot residuals when doing any analysis, but in particular with model selection
That was very useful, thank you! What if I had 2 independent variables(Xs) and 1 dependent variable(Y) and I need to fit a quadratic equation for the data I have?
can u give the vab code to automate this process and the graph is not required i want polynomial equation with 5th degree order and where 5th degree polynomial is {f(x)=(A)+(Bx)+(C(x^2))+(D(x^3))+(E(x^4))+(F(x^5))} where f(x) and x values are give i required the values of A,B,C,D,E&F
Maybe there's a way to change those coefficients automatically, if the polynomial equation is set up as a formula - perhaps also with some conditional formatting or goal seek. I'm not sure, but we can play with it and figure it out!