Page 1 of 1

Interpreting regression

Posted: Wed Sep 09, 2009 8:14 am
by Meg
Hi
I haven’t studied econometrics, new to EViews and not a skilled statistician. An economist helped me running time series regression.

The question is that how can I built 3 equations based on following outputs from final models and interpret them. I can’t interpret the t-1, MA and AR.

First example
Dependent variable: Y
Explanatory variables and their coefficients :
Y t-1, 0.844921
X t, -4.27
Z t, 29.6
Vt, 17917
Constant, -123283
And MA( moving average): MA(1) = -0.984

Second:
Dependent variable: Y
X t-1, 0.69
Z t-1, -3.2
Constant, 7124
AR(1), -0.58
MA(1), 0.99

3rd
Dependent variable d(Y)t
Explanatory variables and their coefficients :
d(Y)t-1, -0.86
constant, 0.53
MA(2), -0.96


Your help is really appreciated.

Re: Interpreting regression

Posted: Wed Sep 09, 2009 8:32 am
by startz
t-1 references the variable the previous period.
AR refers to an autorregressive process
MA refers to a moving average process

Re: Interpreting regression

Posted: Wed Sep 09, 2009 8:38 am
by EViews Gareth
If you don't have any basic time series text books handy, it's probably worth giving Chapter 26 of User Guide II a quick read, it should explain the basics behind those terms.

Re: Interpreting regression

Posted: Wed Sep 09, 2009 8:45 am
by Meg
Thanks STATZ for your quick reply,

For example in Y=0.5X+2.3 we say if X increases by 1 unit, Y will increase by 2.8 unit.

what I need is methematical interpretation of one or more of 3 examples.


Hello QMS , I will read chapter 26 in II.


Regards

Re: Interpreting regression

Posted: Wed Sep 09, 2009 9:05 am
by startz
For example in Y=0.5X+2.3 we say if X increases by 1 unit, Y will increase by 2.8 unit.
Unfortunately, that's wrong.

It sounds like you're starting at the very beginning of learning these tools, as we all did at one time. Perhaps you should back to your helpful economist for some introductory instruction.

Re: Interpreting regression

Posted: Thu Sep 10, 2009 5:36 am
by Meg
Sorry I am confused and did a mistake in the Y=0.5X+2.3: if X increases by one unit, Y will increase by 0.5.

Isn't it possible to give me a clue for of one of my examples. The economist is not avaibale now. I took a look at chapter 26 and couldn't get what I am looking for. I am confused with my limited time. Books give a lot of maths and I can't understand all. I spoke with a statisticain he hasn't experience in time series.

Regards