Opened 13 years ago

Closed 12 years ago

#351 closed request (wontfix)

Estimating parameter p (and confidence interval) in binomial distribution

Reported by: Peter Owned by: Peter
Priority: minor Milestone:
Component: statistics Version: trunk
Keywords: Cc:

Description

related to ticket:258

When performing a permutation test you are sampling from a binary (Bernoulli) distribution and number of successes, k, will follow a binomial distribution. Typically the parameter p is obviously estimated as k/n, but I would like some functionality how to calculate confidence interval.

Should check into GSL to see if there is useful support.

I am not sure how the calculation should be performed but I suppose a Baysian approach might be sound, having a uniform distribution on p.

Change History (2)

comment:2 Changed 12 years ago by Peter

Milestone: yat 0.x+
Resolution: → wontfix
Status: new → closed

This is provided by boost 1.35:

​http://www.boost.org/doc/libs/1_38_0/libs/math/doc/sf_and_dist/html/math_toolkit/dist/stat_tut/weg/binom_eg/binom_conf.html

Note: See TracTickets for help on using tickets.