A Ridge Regression estimator for the zero-inflated Poisson model
The zero inflated Poisson regression model is very common when analysing economic data that comes in the form of non-negative integers since it accounts for excess zeros and over-dispersion of the dependent variable. This model may be used in innovation analysis to see for example the impact on different economic factors has on the number of patents of companies, how frequently mortgages or credit-card fail to meet their financial obligations, the amount of take-over bids firms receives and the number of products different firms export, etc. However, a problem often encountered when analyzing economic data that has not been addressed for this model is multicollinearity. This paper proposes ridge regression estimators and some methods of estimating the ridge parameter k for the non-negative model. A simulation study has been conducted to compare the performance of the estimators. Both MSE and MAE are considered as performance criterion. The simulation study shows that some estimators are better than the commonly applied maximum likelihood estimator and some other ridge regression estimators. Some useful estimators are recommended for the practitioners.
|Date of creation:||18 Oct 2011|
|Date of revision:|
|Contact details of provider:|| Postal: CESIS - Centre of Excellence for Science and Innovation Studies, Royal Institute of Technology, SE-100 44 Stockholm, Sweden|
Phone: +46 8 790 95 63
Web page: http://www.infra.kth.se/cesis/
More information through EDIRC
When requesting a correction, please mention this item's handle: RePEc:hhs:cesisp:0257. See general information about how to correct material in RePEc.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (Vardan Hovsepyan)
If references are entirely missing, you can add them using this form.