Confidence Bounds of Scale Parameters Using Pilot Samples
P. Chiou (2003). Confidence Bounds of Scale Parameters Using Pilot Samples. Journal of Statistical Research, Vol. 37, No. 1, pp. 21-29.
The rst sample in a possible double sampling plan can be referred to as a pilot sample. A pilot study is often done prior to a more thorough study to determine whether further sampling should be conducted. If the second sample is taken, then the data from these two samples can be used for inference purposes. This pa- per studies the condence bounds of scale parameters using this double sampling procedure. The condence bound obtained from the two samples constitutes an exact conditional condence bound (CCB) for the parameter. The conditional coverage probability of the usual condence bound (UCB) is obtained by comput- ing the coverage probability under the conditional probability density function. It is shown that the conditional coverage probability of the UCB is uniformly less than the nominal level.
Download Fulltext (chiou.pdf)
