It is possible to run a regression of both Seconday and primary data in the same model? I mean, when the dependent variable is primary data to be sourced via questionnaire and the Independent variable is secondary data to be gathered from published financial statements?
For Example: if the topic is Capital Budgeting moderator and shareholders wealth (SHW). Capital budgeting moderators is proxy by inflation , management attitude to risk, Economic condition and Political instability. while SHW is proxy by Market value, Profitability and Retained earnings.
Jerry -
I don't know your subject matter or data, but if your question is Can administrative data be used as regressor data? then yes, it can.
Also, for survey data, such "auxiliary" data can be regressor data for prediction, or used in model-assisted design-based sampling and estimation.
The following may be on- or off-base for you, but I'll include it, just in case:
Brewer, K.R.W. (2014), “Three controversies in the history of survey sampling,” Survey Methodology,
(December 2013/January 2014), Vol 39, No 2, pp. 249-262. Statistics Canada, Catalogue No. 12-001-X.
http://www.statcan.gc.ca/pub/12-001-x/2013002/article/11883-eng.htm
Ken Brewer's paper contains an interesting history of the development of survey sampling and estimation.
Särndal, CE, Swensson, B. and Wretman, J. (1992), Model Assisted Survey Sampling, Springer-Verlang.
Even if only tangentially related, you might want to check out the above.
Cheers - Jim
Hi Jerry,
As Dr James has stated, yes the data can be used, but you need to know what methods and transformations (if any) were used on the data before it was used or published.
Regards
I feel it would not be appropriate for a variable to be sourced using primary data (questionnaire) while the other variable in the same model be sourced through secondary data. Reason is this, the primary data to be sourced using the questionnaire will get the reaction of the respondents as at the point in time when the questionnaires were filled by them, the secondary data on the other hand would have been prepared at a prior data. The circumstances prevalent when the secondary data was prepared may have been changed when the respondents are filling the questionnaires. Hence, if this is done, the regression analysis will be carried out using two different forms of data gotten from different prevalent circumstances. This is why, a primary data should be regressed on a primary data, while a secondary data should be regressed against a secondary data. My opinion is however subjected to correction if wrong.
Thanks.
Ajibade Ayodeji. T.
- Badges
- Science topic