2
$\begingroup$

In preparation for my end of semester R exam, I've been working on the following CSV file which was modelled on R, which I'm currently having problems figuring out. The excercise is from the textbook'Applied Statistics: Principles and Examples' by D. David Roxbee Cox.

enter image description here

The data is taken from a real study into Light Water Reactor (LWR) power plants constructed in the USA in the 1970s. The aim of the study was to develop a model that could be used to model the capital cost involved in the construction of LWR power plants. The data set includes 32 LWR power plants and contains the following variables:

The titles PR, NE, CT, BW and PT have outputs that consist of 1 and 0 where 1=Yes and 0=No.

The data set includes 32 LWR power plants and contains the following variables:

enter image description here enter image description here

Now that I have introduced the domain of the question - here is my problem:

I've been asked to 'add suitable terms' to that model in order to investigate whether there is good reason to believe that the impact of a power plant's capacity on construction costs might not be the same for plants that did have a LWR on the same site previously and those that did not.

The summary results of my final model in R is shown below:

enter image description here

Therefore, the equation of my final model is:

C = -541.82231+ 26.11545T1+ 5.22321T2+ 0.41890S- 94.33786PR+108.34622NE- 132.58098PT

I tried fitting a model assessing just the S (power capacity) and PR (whether LWR existed on the same site previously) terms against C (construction cost) but the wording of the question states 'add suitable terms' to that model in order to investigate...' causes me to believe what I wanted to do is wrong because I would have been taking out terms.

Does anyone know what suitable terms I should add and how I test the correlation?

$\endgroup$
4
  • 2
    $\begingroup$ This data set is boot::nuclear -- it comes with R $\endgroup$
    – Glen_b
    Apr 29, 2013 at 8:33
  • 2
    $\begingroup$ I took the liberty to add the “self-study” tag. $\endgroup$
    – Gala
    Apr 29, 2013 at 11:09
  • $\begingroup$ @Glen_b Sorry what does boot::nuclear mean? $\endgroup$
    – methuselah
    Apr 29, 2013 at 12:58
  • 2
    $\begingroup$ try typing it at the R prompt, and see. ("boot" is a package that comes with R; "nuclear" is a data set in the "boot" package; to see what the :: operator does, try ?"::"; note that boot is not loaded by default, otherwise you could just type nuclear) $\endgroup$
    – Glen_b
    Apr 29, 2013 at 23:21

1 Answer 1

2
$\begingroup$

Just a hint, first: The question invites you to look at an interaction. You can keep the other terms in the model (you should in any case keep S and PR) but you do need to add something to test that. Does it make sense?

$\endgroup$
3
  • $\begingroup$ please does anyone know what I need to add to test this? $\endgroup$
    – methuselah
    May 1, 2013 at 12:08
  • $\begingroup$ I don't mind giving you some R code but I understand that the site's policy is to guide toward a solution rather than solve the exercise for you. Where are you stuck? Did you check in your course notes/support material/textbook if you can find some material on interactions? Are you unsure about what an interaction is and why you need to add the corresponding term to the model? Or are you only looking for the right syntax to fit this model in R? $\endgroup$
    – Gala
    May 1, 2013 at 12:32
  • $\begingroup$ Try this. $\endgroup$ May 1, 2013 at 13:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.