1

I used the method pcls in order to make a simple regression (price ~ livingArea) with constraints. I set the constraint for b0 > 1000 but the result I got is very bad.

enter image description here

y = price and X = livingArea

And I think that if the result is bad for a regression with only one explanatory variable, then the result will be worse for a model with one dependent variable and more explanatory variables.

Is it possible that the result was bad because of the constraint I set?

If there is no value for the intercept, can I assume that the value for it is 0?

Paul
  • 471
  • (1) Many answers on site already http://stats.stackexchange.com/search?q=+constrained+regression+r $\qquad \quad:$ (2) it sounds like you only have a single y-variable (response, DV) being price; if so, that would be multiple regression, not multivariate regression; if that's the case please fix your text and your tags accordingly (and if it's not the case please clarify what your response consists of). – Glen_b Jun 08 '15 at 19:40
  • While I understand your premise, I would advise some caution in this. If you are using OLS - or some variant. Then ill fitted varialbes will, at least very often, be the consequence of something in the data that you do not understand (and therefore something you fail to model properly).

    If you add your model, and some generics about the data, maybe we think of some hints/tricks...

     – Repmat Jun 08 '15 at 19:45
  • @user777 I saw that example, but didn't understood it. Could you please explain it. – Paul Jun 08 '15 at 20:09
  • 1
    Your question is already answered on the site. Please read the linked thread (& some other relevant threads that show up upon searching the site). If you still have a question afterwards, come back here & edit your Q to state what you've learned & what you still need to know (nb, 'I didn't understand it' is inadequate). Then we can provide the information you need without simply duplicating material elsewhere that already didn't help you. – gung - Reinstate Monica Jun 08 '15 at 20:18

0 Answers0