Objective: This paper consists on examining a longitudinal data, output of generated data, via constructing a semiparametric model, the wavelets analysis will be applied as smoothing approach for the nonparametric part of the model. Material and Methods: Mixed effects models have been largely examined due to their flexibility in handling data without constraints. Mixed models could be characterized with their parametric and nonparametric features. Indeed, semiparametric mixed models which combine parametric and nonparametric features, started to receive more attention notably for examining longitudinal data. Regarding the nonparametric features, smoothing approaches should be applied. Recently, the wavelets analysis has been considered as a powerful mathematic tool to decompose a series due to its multiresolution features (frequential and temporal) and some researchers mentioned it as a smoothing approach for large data, but the wavelets features for smoothing still not commonly applied on longitudinal data. Results: A data is generated referring to a previous published hypertension study by National Institute Health. The results show that the wavelets analysis has a strong capacity as smoothing approach compared to well-known other smoothing methods; Root Mean Square Errors are calculated, and via the constructed semiparametric model, it has been confirmed that the incident hypertension is related to the high Systolic blood pressure, high diastolic blood pressure and low BMI.
Keywords: Longitudinal data; smoothing approaches; wavelets decomposition; semiparametric model; mixed models
Amaç: Bu makale, türetilmiş bir longitudinal veri seti için, bir yarı parametrik model kurularak incelenmesi üzerine olup, modelin parametrik olmayan kısmı için pürüzsüzleştirme yaklaşımı dikkate alınarak dalga analizi uygulanacaktır. Gereç ve Yöntemler: Karışık etki modelleri, verilerin kısıtlama olmadan kullanılmasındaki esneklikleri nedeniyle yaygın biçimde çalışılmaktadır. Karışık modeller parametrik ve parametrik olmayan özellikleri dikkate alınarak incelenebilir. Nitekim, parametrik ve parametrik olmayan özellikleri birleştiren yarı parametrik karışık modeller, özellikle longitudinal verilerin analizi için daha fazla çalışılmaya başlanmıştır. Parametrik olmayan özellikler ile ilgili olarak, pürüzsüzleştirme yaklaşımları uygulanmalıdır. Son zamanlarda, dalga analizi, çoklu çözünürlük özellikleri (sıklık ve zamansal) nedeniyle bir dizi ayrıştırmak için güçlü bir matematiksel araç olarak kabul edilmektedir, ancak dalga analizi hala longitudinal veri seti için yaygın bir biçimde kullanılmamaktadır. Bulgular: National Institute Health tarafından daha önce yayınlanmış bir hipertansiyon çalışmasına atıfta bulunan bir veri üretilmiştir. Sonuçlar, bilinen diğer pürüzsüzleştirme yöntemlerine kıyasla dalga analizinin, pürüzsüzleştirme yaklaşımı olarak güçlü bir yaklaşım olduğunu göstermektedir; Ortalama Kare Kök Hatalar hesaplanmış ve oluşturulan semiparametrik model aracılığıyla, hipertansiyonunun yüksek Sistolik kan basıncı, yüksek diyastolik kan basıncı ve düşük BMI ile ilişkili olduğu doğrulanmıştır.
Anahtar Kelimeler: Longitudinal boylamsal veri; pürüzsüzleştirme yaklaşımları;dalga ayrışması; yarı parametrik model; karma modeller
- Maxwell SE, Delaney HD, Kelley K. Designing Experiments and Analyzing Data A Model Comparison Perspective. 3rd ed. New York: Routledge; 2018. p.879-81.
- Ruppert D, Wand MP, Carroll RJ. Semiparametric Regression. 1st ed. New York: Cambridge University Press; 2003. [Crossref]
- Szczesniak RD, Li D, Raouf SA. Semiparametric mixed models for medical monitoring data: an overview. J Biom Biostat. 2015;6(2). https://doi. org/10.4172/2155-6180.1000234. [Crossref] [PubMed] [PMC]
- Hulin W, Jin-Ting Z. Semiparametric models. Nonparametric Regression Methods for Longitudinal Data Analysis. Mixed-effects Modeling Approaches. 1st ed. New Jersey: WILEY Series in Probability and Statistics; 2007. p.229-70.
- Ngo L, Wand MP. Smoothing with mixed model software. J Statist Softw. 2004;9(1):1-54. https://doi.org/10.18637/jss.v009.i01. [Crossref]
- Szczesniak RD, McPhail GL, Duan LL, Macaluso M, Amin RS, Clancy JP. A semiparametric approach to estimate rapid lung function decline in cystic fibrosis. Ann Epidemiol. 2013;23(12):771-7. https://doi.org/10.1016/j.annepidem.2013.08.009. [Crossref]
- Zhang D, Lin X. Hypothesis testing in semiparametric additive mixed models. Biostatistics. 2003;4(1):57-74. https://doi.org/10.1093/biostatistics/4.1.57. [Crossref]
- Cleveland W. Robust locally weighted regression and smoothing scatterplots. J Am Stat Assoc. 1979;74(368):829-36. [Crossref]
- Wu CO, Chiang CT. Kernel smoothing on varying coefficient models with longitudinal dependent variable. Stat Sin. 2000;10:433-56.
- Verbyla AP, Cullis BR, Kenward MG, Welham SJ. The analysis of designed experiments and longitudinal data by using smoothing splines. Journal of the Royal Statistical Society Series C-Applied Statistics. 1999;48:269-300. https://doi.org/10.1111/1467-9876.00154. [Crossref]
- Wand MP, Jones MC. Univariate kernel density estimation. Kernel Smoothing. Chapter 2. 1st ed. New York: Chapman & Hall/CRC; 1995. p.10-4.
- Diggle PJ, Heagerty PJ, Liang K, Zeger SL. Analysis of Longitudinal Data. 2nd ed. United Kingdom: Oxford University Press; 2013.
- Fitzmaurice GM, Laird N, Ware HJ. Applied Longitudinal Analysis. 1st ed. New Jersey: John Wiley & Sons; 2004.
- Hedeker D, Gibbons R. Longitudinal Data Analysis. 1st ed. New Jersey: Wiley-Blackwell; 2006.
- Müller HG. Nonparametric Regression Analysis of Longitudinal Data. 1st ed. Berlin Heidelberg New York, London, Paris, Tokyo: Springer-Verlag; 1988. [Crossref]
- Quarta L. Une Introduction (Elémentaire) à La Théorie Des Ondelettes. 2nd ed. Belgium: Mons-Hainaut University; 2001. p.4-18.
- Gençay R, Selçuk F, Whitcher B. An Introduction to Wavelets and Other Filtering Methods in Finance and Economics. Chapters 3, 4 & 5. 1st ed. California; Elsevier, Academic Press; 2002. p.96-234. [Crossref]
- Jin-ting Z. Nonparametric Mixed-Effects Models for Longitudinal Data. South Korea: University of Seoul; 2007.
- Ruppert D, Wand MP, Carroll RJ. Semiparametric Regression. 1st ed. New York: Cambridge University Press; 2003. https://doi.org/10.1017/ CBO9780511755453. [Crossref]
- Lee TCM. Smoothing parameter selection for smoothing splines: a simulation study. Computational Statistics & Data Analysis. 2002;42:139-48. https://doi. org/10.1016/S0167-9473(02)00159-7. [Crossref]
- Eubank RL. Nonparametric Regression and Spline Smoothing. 2nd ed. New York: Marcel Dekker Inc.; 1999.
- Wahba G. Estimating the smoothing parameter. Spline Models for Observational Data. 1st ed. Philadelphia: CBMS-NSF Regional Conference Series in Applied Mathematics; 1990. p.45-65. https://doi.org/10.1137/1.9781611970128. [Crossref]
- Green PJ, Silverman BW. Nonparametric Regression and Generalized Linear Models. 1st ed. London: Chapman & Hall/CRC; 1994. [Crossref]
- Tarter ME, Lock MD. Model-free Curve Estimation. 1st ed. New York, Chapman & Hall/CRC; 1993.
- Ogden RT. Essential Wavelets for Statistical Applications and Data Analysis. 1st ed. Boston, Birkhäuser; 1997. [Crossref]
- Ramsay JO. Kernel smoothing approaches to nonparametric item characteristic curve estimation. Psychometrika. 1991;56(4):611-30. http://doi.org/10.1007/ BF02294494. [Crossref]
- Fan J, Gijbels I. Local Polynomial Modelling and Its Applications. 1st ed. London: Chapman & Hall/CRC; 1996.
- Friedman JH. Multivariate adaptive regression splines (with discussion). Ann Statist. 1991;19(1):1-67. http://doi.org/10.1214/aos/1176347969. [Crossref]
- Stone CJ, Hansen MH, Kooperberg C, Truong YK. Polynomial splines and their tensor products in extended linear modeling. Ann Statist. 1997;25:1371-425. [Crossref]
- Hansen MH, Kooperberg C. Spline adaptation in extended linear models (with discussion). Statist Sci. 2002;17(1):2-51. [Crossref]
- Fan J, Zhang JT. Two-step estimation of functional linear models with applications to longitudinal data. Journal of Royal Statistical Society, Series B 2000;62:303-22. 10. http://doi.org/1111/1467-9868.00233. [Crossref]
- Hoover DR, Rice JA, Wu CO, Yang LP. Nonparametric smoothing estimates of time-varying coefficient models with longitudinal data. Biometrika. 1998;85:809-22. http://doi.org/10.1093/biomet/85.4.809. [Crossref]
- Wu H, Zhang JT. The study of long-term HIV dynamics using semiparametric nonlinear mixed-effects models. Statistics in Medicine. 2002a;21:3655-75. http://doi.org/10.1002/sim.1317. [Crossref]
- Wang N, Carroll RJ, Lin X. Efficient semiparametric marginal estimation for longitudinal /clustered data. Journal of American Statistical Association; 2005; 100:147-57. https://doi.org/10.1198/016214504000000629. [Crossref]
- Brumback B, Rice J. Smoothing spline models for the analysis of nested and crossed samples of curves. Journal of American Statistical Association. 1998;93:961-94. https://doi.org/10.1080/01621459.1998.10473755. [Crossref]
- Wang Y. Mixed-effects smoothing spline ANOVA. Journal of Royal Statistical Society Series B. 1998;93:341-8.
- Wang Y. Smoothing spline models with correlated random errors. Journal of American Statistical Association. 1998;93:341-8. [Crossref]
- Liang H, Wu H, Carroll RJ. The relationship between virologic and immunologic responses in AIDS clinical research using mixed-effects varying-coefficient semiparametric models with measurement error. Biostatistics. 2003;4:297-312. 10. https://doi.org/1093/biostatistics/4.2.297. [Crossref]
- Rice JA, Wu CO. Nonparametric mixed effects models for unequally sampled noisy curves. Biometrika. 2001;57:253-9. https://doi.org/10.1111/j.0006341X.2001.00253.x. [Crossref]
- Wu H, Zhang JT. Local polynomial mixed-effects models for longitudinal data. Journal of American Statistical Association. 2002;97(459):883-97. https:// doi.org/10.1198/016214502388618672. [Crossref]
- Nason GP, Silverman BW. The discrete wavelet transform in S. J Comput Graph Stat. 1994;3(2):163-91. https://doi.org/10.1198/1061860031671. [Crossref]
- Han M, Liu Y, Xi J. Guo W. Noise smoothing for nonlinear time series using wavelet soft threshold. IEEE Signal Process Lett. 2007;14(1):62-5. https://doi. org/10.1109/LSP.2006.881518. [Crossref]
- Horgan GW. Using wavelets for data smoothing: a simulation study. J Appl Stat. 1999;26(8):923-32. https://doi.org/10.1080/02664769921936. [Crossref]
- Schimek M. Smoothing and Regression Approaches, Computation, and Application. 1st ed. USA: John Wiley & Sons, Inc.; 2000. https://doi. org/10.1002/9781118150658. [Crossref] [PMC]
- Nason GP, Silverman BW. The stationary wavelet transform and some statistical applications. Wavelets and Statistics. 1995;281-300. [Crossref]
- Masset P. Analysis of Financial Time-Series Using Fourier and Wavelet Methods. SSRN. 2008 ;1-36. https://doi.org/10.2139/ssrn.1289420. [Crossref]
- Polikar R. The wavelet tutoriel. Index to Series of Tutoriels to Wavelet Transform. 2nd ed. New Jersey: Rowan University; 2006. p.2-15.
- Conway P, Frame D. A spectral analysis of New Zealand output gaps using Fourier and wavelet techniques. Discussion Paper Series. 2000;1-20.
- Debdas S, Quereshi MF, Reddy A, Chandrakar D, Pansari D. A Wavelet based multiresolution analysis for real time condition monitoring of AC machine using vibration analysis. International Journal of Scientific & Engineering Research 2011;2(10).
- Grossmann A, Morlet J. Decomposition of Hardy functions into square integrable wavelets of constant shape. SIAM Journal of Mathematical Analysis and Applications. 1984;15(4):723-36. https://doi.org/10.1137/0515056. [Crossref]
- Andreopoulos Y, Schaar MVD. Incremental Refinement of Computation for the Discrete Wavelet Transform IEEE Transactions on Signal Processing. 2008;56(1):140-57. https://doi.org/10.1109/TSP.2007.906727. [Crossref]
- Equitz WHR, Cover TM. Successive Refinement of Information. IEEE Transactions on information theory. 1991;37(2). https://doi.org/10.1109/ ITW.1989.761420. [Crossref]
- Graps A. An Introduction to Wavelets. IEEE Computational Science and Engineering. 1995;2(2):50-61. https://doi.org/10.1109/99.388960. [Crossref]
- Chen J, Stock S, Deng C. Sample size estimation through simulation of a random coefficient model by using SAS abstract. PharmaSUG. 2008;3.
- Psioda M. Random Effects Simulation for Sample Size Calculations Using SAS. 1st ed. Carolina: University of North Carolina, Chapel Hill NC; 2012. p.1-11.
.: Process List