Hopefully somebody will be able to shed some light on my SPSS problems!

I have been given 65 values. 57 of these data values are quarterly results and 8 are the holdback data to be used.

I have to do: - Regression with Dummy variables with a linear trend cycle component

Does anyone know what to do as my results aren't making much sense?

For the first part - I obviously split the data into dummy variables for the relevant quarters (Q1-Q4).

I then performed regression analysis - linear. But all my values are extremely large and not significant. Also Q2 has been listed as 'excluded variables' in the results? I have followed the steps and I am unsure why this has happened.

Then I thought of removing Q4, due to multi-collinearity but again the values are still quite large (>.450).

Not sure if I am doing something wrong at the start (especially with the excluded variables aspect)

Anybody got any idea? This is driving me nuts

Update: It won't let me comment back on the main page for some reason.

The data set was given to us: "It is a quarterly series of total consumer lending. It is not seasonally adjusted. The first 57 data values for modelling and choose the remaining 8 data values as holdback data to test your models."

The data is: (last 8 are holdback data) 16180 17425 17424 17240 18240 19880 20143 20545 22155 23344 23717 23467 25442 27278 27848 25704 28919 30280 32095 31041 33182 35067 35557 34420 35948 38643 39612 39185 40143 40056 41360 41343 43652 44554 47903 46460 49402 50254 50335 48763 51529 53481 53482 53882 55219 56180 56037 54106 54915 54641 53805 52179 52026 51522 51733 50672 50882 50878 52199 50261 49615 47995 45273 42836 43321

It has to be SPSS generated.

Email primarybeing12@hotmail.co.uk - not letting me respond to people. Thanks for any help!

Doing the ARIMA forecasting is the next step (which I understand). I have to do regression on the linear/non-linear for this question

If I was to use time, time^2, Q1, Q2, Q3 + lagged variables.

Would I use lagged variables 1-3? Also, I understand the rest, but what benefit does using lagged variables do? As I said, feel free to e-mail me if you can.

  • $\begingroup$ Is this homework? If so, please add the self-study tag. Also what is a "linear trend cycle component"? Plus please tell us what data you have. Clearly you have more than "65 values"; you must have data on season, and probably other things as well. $\endgroup$ – Peter Flom Nov 22 '13 at 19:55
  • $\begingroup$ Is your question rather from "time-series" realm than "regression" realm? $\endgroup$ – ttnphns Nov 22 '13 at 20:05
  • $\begingroup$ Ordinary least squares regression won't work here becuase it is a time series dataset, you can try any time series regression methods in SPSS. $\endgroup$ – forecaster Nov 22 '13 at 20:12
  • $\begingroup$ Can you post the data online ? $\endgroup$ – forecaster Nov 22 '13 at 20:15
  • $\begingroup$ clearly in plotting the data, I see a trend-cycle and seems to be quadratic in nature, you should be using time series regression for this type of problem. If you want to use regression, then you can regress lending = Q1 + Q2 + Q3 + time + timesq, where timesq = time ^2. This would still not give a good fit, you also might want to add lag terms. Also you can try ARIMA in SPSS, that would automatically do this for you. $\endgroup$ – forecaster Nov 22 '13 at 21:46

You might want to do a time series regression as opposed to OLS regression. SAS has a procedure called Proc UCM (unobserved components model) which would do exactly what you are asking for . But I'm not sure if SPSS has a similar procedure. Please check this website for a"trend cycle" example using UCM. http://support.sas.com/rnd/app/examples/ets/melanoma/


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.