Estimation of the percentile of Birnbaum-Saunders distribution and its application to PM2.5 in Northern Thailand

View article
Environmental Science

Main article text

 

Introduction

Methods

where α is the shape parameter and β is the scale parameter.

where zp=Φ1(p) is the standard normal p-th quantile.

Generalized confidence interval approach

where β1 and β2 are two solutions for β can be derived by solving Aβ22Bβ+C=0, A=(n1)J21nLT2, B=(n1)IJ(1IJ)T2, C=(n1)I21nKT2, I=1nni=1Xi, J=1nni=11Xi, K=ni=1(XiI)2, L=ni=1(1XiJ)2, and Tt(n1).

where Rβ is defined in Eq. (3) and s1 and s2 are the observed values of S1 and S2, respectively.

where zp=Φ1(p) is the standard normal p-th quantile, Rβ is defined in Eq. (3), and Rα is defined in Eq. (4).

where Rθ(γ/2) and Rθ(1γ/2) denote the 100(γ/2)-th and 100(1γ/2)-th percentiles of Rθ, respectively.

Bootstrap approach

and

and

where k=1,2,...,B.

where zp=Φ1(p) is the standard normal p-th quantile, ~αk is defined in Eq. (9), and ~βk is defined in Eq. (10).

where ˆθk(γ/2) and ˆθk(1γ/2) denote the 100(γ/2)-th and 100(1γ/2)-th percentiles of ˆθk, respectively.

Bayesian approach

where r is a constant and p(|x) is defined by using Eq. (13). Therefore, β=vur has density p(β|x)p(β|x)dβ. To sample random data points in A(r), the random variables (u,v) are generated from a uniform distribution over a one-dimensional bounded rectangle [0,a(r)]×[b(r),b+(r)].

and

where zp=Φ1(p) is the standard normal p-th quantile.

where θBaye(γ/2) and θBaye(1γ/2) denote the 100(γ/2)-th and 100(1γ/2)-th percentiles of θBaye, respectively.

Highest posterior density approach

where Lθ.HPD and Uθ.HPD are determined using the hdi function within the HDInterval package of the R software suite.

Results

Empirical application

Discussion

Conclusion

Supplemental Information

Daily PM2.5 levels data in Mae Hong Son and Lampang provinces.

Source: Pollution Control Department, Thailand. http://air4thai.pcd.go.th/webV3/#/History.

DOI: 10.7717/peerj.17019/supp-1

R code to construct histograms of PM2.5 data from (A) Mae Hong Son Province (B) Lampang Province.

DOI: 10.7717/peerj.17019/supp-2

R code to construct confidence intervals from all methods in this article.

DOI: 10.7717/peerj.17019/supp-3

Additional Information and Declarations

Competing Interests

The authors declare that they have no competing interests.

Author Contributions

Warisa Thangjai conceived and designed the experiments, performed the experiments, prepared figures and/or tables, authored or reviewed drafts of the article, and approved the final draft.

Sa-Aat Niwitpong analyzed the data, authored or reviewed drafts of the article, and approved the final draft.

Suparat Niwitpong conceived and designed the experiments, analyzed the data, prepared figures and/or tables, authored or reviewed drafts of the article, and approved the final draft.

Data Availability

The following information was supplied regarding data availability:

The data and R code for computing coverage probability and average length of all confidence intervals is available in the Supplemental Files.

Funding

This research was funded by the King Mongkut’s University of Technology North Bangkok. Grant No: KMUTNB-67-KNOW-09. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

4 Citations 717 Views 74 Downloads

MIT

Your institution may have Open Access funds available for qualifying authors. See if you qualify

Publish for free

Comment on Articles or Preprints and we'll waive your author fee
Learn more

Five new journals in Chemistry

Free to publish • Peer-reviewed • From PeerJ
Find out more