Calculate probability for failure (PD) for people who joined research in different time

[Ronein]

Hello,

There is a research of checking failures occurances of people.

The information includes the following fields:

Entrance time (in YYMM format)

Nr_people (Number of people enter the research in this month)

time0_to_time1 (Number of people who fail between time 0 and time 1 since they enter  the research)

time1_to_time2 (Number of people who fail between time 1 and time 2 since they enter  the research )

time2_to_time3 (Number of people who fail between time 2 and time 3 since they enter  the research )

and so on.

As you can see for each group of people there is a different number of months that they were exposed for failure.

For example:

people who joined in 2101 are exposed for failure during 6 months.(till 2106)

people who joined in 2101 are exposed for failure during 5 months.(till 2106)

and so on.

I want to calculated the cumulative probability of failure in this research.

I know to calculate it for each group (time entrance)

but the question is how to calculate it for the whole people in the research.

[Seadrago]

I think this may have something to to with exposure adjusted incidence rates or exposure adjusted event rates. Since each group was exposed at different times you'll need an overall amount of exposure ( denominator usually done in person-years this is all subjects exposure time summed together both failure and success). The numerator would then be the number of people who failed times amount of exposure they had.

[Ronein]

Thanks
It is interesting for me to see your calculation in sas or excel I order to understand the methodology of calculate it

[sbxkoenk]

Hello,

If you enter :

exposure adjusted incidence rates exposure adjusted event rates

as search terms on :

https://www.lexjansen.com/

, you will already get a few hints.

Thanks,

Koen