I have been struggling with this problem the last couple days, whenever I run my selection model, the IMR obtained is equal to NULL. Does anyone know how I can adjust my code to obtain a value? Thank you very much.
Variables used:
EarlyFunding - binary, whether a start-up received early funding or not
NumberofFounder - numeric, how many founders the start-up has
PriceUSD - numeric, start-up acquisition price
CapitalAllocated - numeric, how much capital was allocated to the start-up
DaysActive - numeric, days active before acquisition
MicroSize, SmallSize, LargeSize, XLargeSize - dummy variables, either 1 or 0
NumberofFunding - numeric, how many funding rounds the start-up underwent
My code in:
heckit <- selection(EarlyFunding ~ NumberofFounder,
log(PriceUSD) ~ log(CapitalAllocated) + DaysActive + MicroSize + SmallSize + LargeSize + XLargeSize + NumberofFunding + EarlyFunding)
summary(heckit)$Mills
Console Output:
Tobit treatment model (switching regression model)
Maximum Likelihood estimation
Newton-Raphson maximisation, 4 iterations
Return code 8: successive function values within relative tolerance limit (reltol)
Log-Likelihood: -1805.498
816 observations: 470 non-participants (selection 0) and 346 participants (selection 1)
13 free parameters (df = 803)
Probit selection equation:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.56529 0.09326 -6.062 0.00000000207 ***
NumberofFounder 0.17500 0.03455 5.066 0.00000050558 ***
Outcome equation:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 19.178626134 0.212762493 90.141 < 0.0000000000000002 ***
log(CapitalAllocated) -0.687235433 0.026411267 -26.021 < 0.0000000000000002 ***
DaysActive 0.000006998 0.000025479 0.275 0.783647
MicroSize -2.242781147 0.206837357 -10.843 < 0.0000000000000002 ***
SmallSize -1.131765568 0.109541607 -10.332 < 0.0000000000000002 ***
LargeSize 0.498545219 0.136097816 3.663 0.000265 ***
XLargeSize 1.024228327 0.109721965 9.335 < 0.0000000000000002 ***
NumberofFunding 0.226500314 0.017561332 12.898 < 0.0000000000000002 ***
EarlyFunding 1.648318372 0.287864038 5.726 0.0000000145 ***
Error terms:
Estimate Std. Error t value Pr(>|t|)
sigma 1.36322 0.08403 16.224 < 0.0000000000000002 ***
rho -0.70534 0.08625 -8.178 0.00000000000000112 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
--------------------------------------------
> summary(heckit)$Mills
NULL
>