rado.pl Wrote: ------------------------------------------------------- > it was 0.85 and you shouldn’t add intercept. what > they asked in the question was how dependent > variable would change if we change one independent > by 0,5 HOLDING OTHER VARIABLES CONSTANT. > > corect answer is b1*0,5. that is what definition > of slope coefficient in linear regression model > suggests. This all comes back to what EXACTLY was the question. Did it ask ‘how would the dependent variable change given (blank) and holding all else constant.’ [no intercept] or did it ask ‘what is the change in exchange rates given the following assumptions.’ [full model] It was the morning session and I certainly don’t remember. It could be either as far as I know. I could barely remember my own phone number after that afternoon session. I do, however, plainly remember it saying ‘holding all else constant.’ That said, I expect to miss 4-5 little questions like this. It seems they like to test comprehension, diligence, whatever. IMO, if you include this phrase taken verbatim from the text on calculating the impact of an individual independent variable but you are really asking to find the value of the dependent variable… well, that is a little overboard. You want to be difficult, ask a difficult quant question. I am fine with it either way. I would just prefer difficult concepts over difficult wording.
I’m pretty confident the question asked for the %change in the exchange rate between the $ and AUSD. I actually remember looking for a log-log regression since dependent and independent variables were interpreted as %change…
The question asked for the change in the “exchange rate variable”. I took that to mean the change in the dependent variable, not the change in the exchange rate, so I did not include the intercept. It was an extremely poorly worded question and I could see it being scored either way. But the fact that they included the word “variable” made me believe that it was the change in the dependent variable that they were looking for. If they were just looking for the change in the exchange rate they would not have included the word “variable.” Anyone else have the same thought?
I don’t recall it saying variable.
I feel I have to go -1 for me here based on what Mark has posted previously and the fact that the dependent variable was a change in exch rates. Again, I cannot recall if it said dependent variable or exchange rate variable or change in exchange rate… My recollection at this point is as stated… what is the impact on {something} given an X1 of {something} and holding all else constant. To me that screamed, what is the impact on Y of a change in X1 holding all other X’s constant. **Edit: I recall saying to myself, “self, if you are asked to calc Y you need to include values for all X’s you don’t hold some X’s constant and then choose just one to plug in…” …but whatever it was, we will most likely never see it again or know for sure.