I am modeling the time to start-up outcomes (e.g., new firm or quitting) and exploring how the independent variables of the %internal, debt, and equity influence the time to different start-up outcomes. Obviously endogeneity is an issue, % of each kind of financing depends on the quality of the start-up.
Is there something similar to IVREG2 or 2SLS that can be used via syntax in the stcrreg command context?
One potential option is to estimate models for finance use and save the predicted outcome for use as a latent variable in your time to start equation. Or maybe use group human capital and financial capital deletion tests ex post after estimating your time to start model. Or use 2SLS on the assumption that human capital is largely predetermined at the point of requesting finance then time to start as your second model. Also think about the fact that equity finance decisions are largely dependent on human capital of the founding team whilst debt finance is much more related to collateral availability and prior banking behaviour. Hope this helps? Cheers, Marc
Thanks Marc, good points will try some of the suggestions out and compare model fit.
- Badges
- Science topic
- Similar topics
- Medicine
- Physiology
- Biosignals
- Biomedical Signal Processing
- Statistical Analysis