How to manually change "Adj S^2" value in MackChainLadder?

[LostTriangles]

Hello,

When using the Mack Method in excel, we get #div/0! errors if the tail portion of our triangle sees no development for several periods on all relevant AYs. As can be seen from the picture below, I could hard-code to a 0 in the highlighted cell as a workaround.

However, after playing around with the tail.se and tail.sigma functions in ChainLadder, I've not been able to figure out how I would hard-code something in a similar fashion.

Is there such a work around for triangles with this type of development when running a MackChainLadder process?

Thanks again in advance.

[mages]

Sorry, I am lost here. Reproducible in R code please, not Excel screenshots.

On 16 Aug 2016, at 17:41, LostTriangles [email protected] wrote:

Hello,

When using the Mack Method in excel, we get #div/0! errors if the tail portion of our triangle sees no development for several periods on all relevant AYs. As can be seen from the picture below, I could hard-code to a 0 in the highlighted cell as a workaround.

However, after playing around with the tail.se and tail.sigma functions in ChainLadder, I've not been able to figure out how I would hard-code something in a similar fashion.

Is there such a work around for triangles with this type of development when running a MackChainLadder process?

Thanks again in advance.

https://cloud.githubusercontent.com/assets/20261406/17707417/5688133c-63ae-11e6-9f92-b7f69a54b2f1.PNG
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub #18, or mute the thread https://github.com/notifications/unsubscribe-auth/ABF-1f6aNJAwr6C46bnyD45RXffh4D6cks5qgegugaJpZM4JlnLT.

[LostTriangles]

My issue has less to do with R code than it does with asking for the R code to change a specific entry in Mack Chain Ladder.

For example, if I were to import the triangle from the screen shot above and ran something like

R> mack<-MackChainLadder(sample_tri, est.sigma = "Mack")
R> mack

I would receive NaN errors for the Standard Error and CV(IBNR) columns due to the nature of our triangle (there is no development for each AY in the last 4 periods). In excel format, I'd normally just hardcode the "Adjusted S^2" value to 0 when an error occurs due to having 0s from the prior 2 Adjusted S^2 values.

I'd like to be able to do the same in R. I understand one workaround would be to simply cut off the triangle at AY 3 but I'd like to incorporate some of the variability from AYs 1-3 in the earlier development periods.

Thank you

[chiefmurph]

Dear Lost,
We applaud your urgency and your enthusiasm! However, one "rule" of open-source contributing is that you must supply a small, reproducible example. It's up to you to generate the data that drives the offensive result. Then it's up to us to respond.
In your favor, we are not unfamiliar with the potential for your issue. Looking forward to the opportunity of discussing your example.
Thank you.