calculating IRR (Internal Rate of Return)

[Guy Lakan]

hi,

my data looks as folow:

id	year	flow
1	2008	-100,000
1	2010	30,000
1	2012	40,000
1	2013	35,000

I want to generate (maybe egen) a new var which will include the IRR of each project (id). I have two main problems:

1. As you can see there is no observations for some of the years (its fine because the flow in these years was zero) but it is problematic for the irr calculation. How could I add observation for the missing years and zero for the flow?

id	year	flow
1	2009	0
1	2011	0

2. What is the best way to calculate the IRR?

Thank you,
Guy.

[Andrew Musau]

* Example generated by -dataex-. To install: ssc install dataex
clear
input byte id int year long flow
1 2008 -100000
1 2010   30000
1 2012   40000
1 2013   35000
end

gen new=0
xtset id year
tsfill
replace flow=0 if new
drop new

Res.:

. l, sepby(id)

     +---------------------+
     | id   year      flow |
     |---------------------|
  1. |  1   2008   -100000 |
  2. |  1   2009         0 |
  3. |  1   2010     30000 |
  4. |  1   2011         0 |
  5. |  1   2012     40000 |
  6. |  1   2013     35000 |
     +---------------------+

2. What is the best way to calculate the IRR?

On calculating IRR, the following community contributed command may help, but I have never used it.

https://ideas.repec.org/c/boc/bocode/s457597.html

[Guillaume Daudin]

Hi,
I have a similar issue except that :

1) I have more than one negative flow in my cash flow chronology
2) Some investment are net losers, so the net current value is negative.

In that case, IRR by Maximo Sangiacomo does not work.

Any idea ? I could use the Solver on Excel, but I hope there is a better way to do it !
Thanks,
Guillaume