[Feature] Add a Life extractor

[lrnv]

No description provided.

[lrnv]

I have an issue that is that

julia> L = Life(slopop[:male], 7000, 1940*365.241)
Life(
∂t: [5.093170329928398e-11, 304.8199999999488, 60.42100000005121, 304.8199999999488, 60.42100000005121, 304.8199999999488, 60.42100000005121, 304.8199999999488, 60.42100000005121, 304.8199999999488  …  60.42100000005121, 304.8199999999488, 60.42100000005121, 304.8199999999488, 60.42100000005121, 304.8199999999488, 60.42100000005121, 304.8199999999488, 60.42100000005121, 304.8199999999488]
λ: [1.18259308900975e-5, 1.18259308900975e-5, 1.31461386213319e-5, 1.31461386213319e-5, 1.43568885285431e-5, 1.43568885285431e-5, 1.52653024528616e-5, 1.52653024528616e-5, 1.59261564872642e-5, 1.59261564872642e-5  …  0.000952040196524071, 0.000733071745918347, 0.000660281996864776, 0.00105021724521249, 0.00123749831958367, 0.00112796698379798, 0.000624953529261446, 0.00105591234503242, 0.00111013031972907, 0.000734484865047132]       
)


julia> expectation(L)/265.241
67.35108708597433

which is OK, but

julia> L = Life(slopop[:male], 7000, 1990*365.241)
Life(
∂t: [3.637978807091713e-12, 304.81999999999607, 60.421000000003914, 304.81999999999607, 60.421000000003914, 304.81999999999607, 60.421000000003914, 304.81999999999607, 60.421000000003914, 304.81999999999607  …  60.421000000003914, 304.81999999999607, 60.421000000003914, 304.81999999999607, 60.421000000003914, 304.81999999999607, 60.421000000003914, 304.81999999999607, 60.421000000003914, 304.81999999999607]
λ: [5.5910580526963e-6, 3.17782866861003e-6, 3.45195218341532e-6, 4.76812723189004e-6, 4.60357074517423e-6, 3.39712528476538e-6, 3.45195218341532e-6, 4.1922227932534e-6, 5.20699288852109e-6, 4.21964406742172e-6  …  4.38476258334695e-6, 5.32225320572007e-6, 7.32946027661339e-6, 6.63278668692437e-6, 7.52554010487343e-6, 4.17886868465e-6, 8.4691237966412e-6, 6.39964683932395e-6, 9.31110361684609e-6, 9.953004363499262e-6]
)


julia> expectation(L)/265.241
41.55705546700694

i.e. people youger have a life expectation that goes up to today and then they are out.
maybe we need to do something w.r.t. the remaining survival at the end of tha table ? This might be one of the reason our numbers do not reconciles with R. But then the question is how do they correct for this in R ?

Edit : 265 instead of 365 but same conclusion

[lrnv]

With the last modifications this should be good to go. Todo before merging :

• Document the functionality correctly. Basically it allows to extract a life (given covaraites) from a ratetable and then use this life as a random variables: survival, expectation, random number generation, etc... We should add doctrings but also add a part on the online docs about the functionality.
• Merge and register a new version fo the package.

[codecov]

Codecov Report

Attention: Patch coverage is 77.19298% with 13 lines in your changes are missing coverage. Please review.

Project coverage is 73.68%. Comparing base (136610f) to head (c041c48).
Report is 6 commits behind head on main.

Files	Patch %	Lines
src/Life.jl	77.19%	13 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main       #8      +/-   ##
==========================================
+ Coverage   70.17%   73.68%   +3.50%     
==========================================
  Files           3        4       +1     
  Lines          57      114      +57     
==========================================
+ Hits           40       84      +44     
- Misses         17       30      +13     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[lrnv]

Yep, looks good to me. However it lacks a test to be good to go. Only a test that enforces the syntax of calling it.

[lrnv]

let me write up the test quickly

[lrnv]

@rimhajal If online tests passes, this should be good to go. As soon as it is merged, we have to register the new version