Abstract
The objective of this research is to investigate the problem of goodness-of-fit testing based on nonparametric density estimation with a data-driven smoothing parameter. The first proposed test statistic λ[α] is itself a smoothing parameter which is selected to minimize an estimated MISE for a truncated series estimator of the comparison density function. Therefore, this test statistic leads immediately to a point estimate of the density function in the event that H[0] is rejected. The limiting distribution of λ[α] was obtained under the null hypothesis. It was also shown that this test is consistent against fixed alternatives. The other new test statistic is essentially a Neyman smooth test that uses an estimated smoothing parameter to choose the number of term s in the statistic. In our simulation study, we found this test to have excellent empirical power properties.
Kim, Jong-Tae (1992). Testing goodness-of-fit via order selection criteria. Texas A&M University. Texas A&M University. Libraries. Available electronically from
https : / /hdl .handle .net /1969 .1 /DISSERTATIONS -1420108.