[R] binequality - Problems with fitFunc()

[Tony Greening]

binequality looks like just the package I need, specifically for obtaining Gini income coefficients for binned data.


I have upgraded to the latest version of R, the latest version of RStudio, and updated all packages. Still I have a persistent problem using fitFunc().


To keep things simple, I am testing things using the example code in the binequality manual, page 6 (or, "?fitFunc").

However, it exhibits the same problems as does my code and another example found elsewhere on the Internet, namely NA results for all fields of interest, as follows:




Time difference of 0.776551 secs
for LNO fit across 2 distributions

$datOut
       State obsMean distribution estMean var cv cv_sqr gini theil MLD
1 California      NA          LNO      NA  NA NA     NA   NA    NA  NA
2      Texas      NA          LNO      NA  NA NA     NA   NA    NA  NA
  SDL aic bic didConverge logLikelihood nparams median sd
1  NA  NA  NA       FALSE            NA      NA     NA NA
2  NA  NA  NA       FALSE            NA      NA     NA NA

$timeStamp
[1] 1497327774

$parameters
           mu sigma nu tau
California NA    NA NA  NA
Texas      NA    NA NA  NA

$quantiles
NULL



To summarise, I have tried fitFunc() in my own code and it produced similar results to that above. I found another example (for a different purpose) and tried it, and it also gave these results. Finally, trying the example in the fitFunc example code proves that the results are (i) consistent and (ii) not what any of the three authors would be expecting.


Any thoughts?



	[[alternative HTML version deleted]]