Re: TESTING HYPOTHESES BY MONTE CARLO



Tomsky says: samples don't have confidence intervals, parameters do

Luis Amaral Afonso says: Tomsky is wrong

Luis: here is a sample, give me the confidence inteval for this sample

4.5 7.6 8.4 3.3 4.2 9.2 1.9 10.0

Can't do it without reference to a parameter, can you?

Once again, Tomsky is right and you are wrong.

BWAHAHAHAHA


John Smith
.



Relevant Pages

  • Adumbocito rolls again!
    ... According to Luis Amaral Afonso, he can use Monte Carlo to compute confidence intervals for samples, without needing parameters. ... Tomsky says: samples don't have confidence intervals, ...
    (sci.stat.math)
  • Re: Jarque-Bera test: confidence intervals for normal data
    ... Jack L. Tomsky KNOWS NOTHING about it is saying. ... what is the parameter for which you obtained your confidence intervals? ...
    (sci.stat.math)
  • Re: CONFIDENCE LIMITS AROUND A PERCENTILE
    ... Statistics) dare to oppose what Conover treats this ... Luis Amaral Afonso ... CONFIDENCE INTERVALS FOR QUANTILES ... This algorithm is based on the Normal Distribution ...
    (sci.stat.math)
  • Re: I denand an immediate explanation to Jack Tomsky!
    ... sample parameters. ... In the rests above we standardise data in order to ... Luis Amaral Afonso ... I'm merely accepting you at your word and have found that your confidence intervals have such poor properties that they should never be used. ...
    (sci.stat.math)
  • Re: I denand an immediate explanation to Jack Tomsky!
    ... Monte Carlo usefulness to test hypotheses. ... Luis Amaral Afonso ... you're the one who called them "confidence intervals for normal data". ...
    (sci.stat.math)