×

Pattern recognition of longitudinal trial data with nonignorable missingness: an empirical case study. (English) Zbl 1187.68448

Summary: Methods for identifying meaningful growth patterns of longitudinal trial data with both nonignorable intermittent and drop-out missingness are rare. In this study, a combined approach with statistical and data mining techniques is utilized to address the nonignorable missing data issue in growth pattern recognition. First, a parallel mixture model is proposed to model the nonignorable missing information from a real-world patient-oriented study and concurrently to estimate the growth trajectories of participants. Then, based on individual growth parameter estimates and their auxiliary feature attributes, a fuzzy clustering method is incorporated to identify the growth patterns. This case study demonstrates that the combined multi-step approach can achieve both statistical generality and computational efficiency for growth pattern recognition in longitudinal studies with nonignorable missing data.

MSC:

68T10 Pattern recognition, speech recognition

Software:

Mplus
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] DOI: 10.1177/0049124100028003003 · doi:10.1177/0049124100028003003
[2] DOI: 10.1080/01621459.1957.10501379 · doi:10.1080/01621459.1957.10501379
[3] Dempster A. P., J. Royal Stat. Soc. Series B Methodol. 39 pp 1–
[4] DOI: 10.1093/biomet/72.3.497 · Zbl 0609.62082 · doi:10.1093/biomet/72.3.497
[5] Little R. J. A., Statistical Analysis with Missing Data (1987) · Zbl 0665.62004
[6] DOI: 10.1002/9781119013563 · doi:10.1002/9781119013563
[7] DOI: 10.1093/biomet/80.2.267 · Zbl 0778.62022 · doi:10.1093/biomet/80.2.267
[8] DOI: 10.1191/096228099671525676 · doi:10.1191/096228099671525676
[9] DOI: 10.2307/2986113 · Zbl 0825.62010 · doi:10.2307/2986113
[10] DOI: 10.1080/01621459.1995.10476615 · doi:10.1080/01621459.1995.10476615
[11] DOI: 10.1093/biomet/63.3.581 · Zbl 0344.62034 · doi:10.1093/biomet/63.3.581
[12] DOI: 10.1111/j.1541-0420.2005.00340.x · Zbl 1078.62073 · doi:10.1111/j.1541-0420.2005.00340.x
[13] DOI: 10.1080/01621459.1952.10483446 · doi:10.1080/01621459.1952.10483446
[14] Little R. J. A., Int. Stat. Rev./Revue Internationale de Statistique 54 pp 139–
[15] DOI: 10.1093/biomet/70.1.41 · Zbl 0522.62091 · doi:10.1093/biomet/70.1.41
[16] Rosenbaum P. R., Am. Stat. 39 pp 33–
[17] DOI: 10.1080/01621459.1999.10473870 · doi:10.1080/01621459.1999.10473870
[18] DOI: 10.1002/sim.4780121807 · doi:10.1002/sim.4780121807
[19] DOI: 10.1080/01621459.1995.10476493 · doi:10.1080/01621459.1995.10476493
[20] DOI: 10.2307/2531734 · Zbl 0715.62136 · doi:10.2307/2531734
[21] DOI: 10.1002/sim.4780110608 · doi:10.1002/sim.4780110608
[22] Afifi A. A., J. Am. Stat. Assoc. 61 pp 595–
[23] Buck S. F., J. Royal Stat. Soc. Series B (Methodological) 22 pp 302–
[24] Marker D. A., Survey Nonresponse (2002)
[25] DOI: 10.1080/01621459.1994.10476768 · doi:10.1080/01621459.1994.10476768
[26] DOI: 10.1080/01621459.1996.10476909 · doi:10.1080/01621459.1996.10476909
[27] DOI: 10.1093/biomet/79.4.811 · Zbl 0764.62008 · doi:10.1093/biomet/79.4.811
[28] DOI: 10.1080/01621459.1998.10473733 · doi:10.1080/01621459.1998.10473733
[29] DOI: 10.1080/01621459.1993.10476366 · doi:10.1080/01621459.1993.10476366
[30] Raudenbush W., Hierarchical Linear Models: Applications and Data Analysis Methods (2002)
[31] DOI: 10.1080/01621459.1996.10476908 · doi:10.1080/01621459.1996.10476908
[32] DOI: 10.1080/01621459.1986.10478280 · doi:10.1080/01621459.1986.10478280
[33] DOI: 10.1037/1082-989X.6.4.330 · doi:10.1037/1082-989X.6.4.330
[34] DOI: 10.1109/TKDE.2003.1245289 · Zbl 05109777 · doi:10.1109/TKDE.2003.1245289
[35] DOI: 10.1080/713827170 · Zbl 05387179 · doi:10.1080/713827170
[36] DOI: 10.1109/TKDE.2005.188 · Zbl 05110092 · doi:10.1109/TKDE.2005.188
[37] Little R. J. A., J. Am. Stat. Assoc. 88 pp 125–
[38] DOI: 10.1093/biomet/81.3.471 · Zbl 0816.62023 · doi:10.1093/biomet/81.3.471
[39] DOI: 10.2307/2533148 · Zbl 0875.62273 · doi:10.2307/2533148
[40] DOI: 10.1037/1082-989X.7.2.147 · doi:10.1037/1082-989X.7.2.147
[41] Verbeke G., Linear Mixed Models for Longitudinal Data (2000) · Zbl 0956.62055
[42] DOI: 10.1207/s15327906mbr2904_1 · doi:10.1207/s15327906mbr2904_1
[43] DOI: 10.1037/1082-989X.2.1.64 · doi:10.1037/1082-989X.2.1.64
[44] DOI: 10.1007/BF02294365 · Zbl 0627.62066 · doi:10.1007/BF02294365
[45] Muthén L., Mplus User’s Guide (2006)
[46] B. Muthén, New Developments and Techniques in Structural Equation Modeling, eds. G. A. Marcoulides and R. E. Schumacker (Lawrence Erlbaum Associates, Mahwah, NJ, 2001) pp. 1–33.
[47] DOI: 10.1037/10409-010 · doi:10.1037/10409-010
[48] DOI: 10.1007/978-1-4899-4541-9 · doi:10.1007/978-1-4899-4541-9
[49] DOI: 10.1207/S15327906MBR3703_3 · doi:10.1207/S15327906MBR3703_3
[50] DOI: 10.1007/978-1-4757-0450-1 · doi:10.1007/978-1-4757-0450-1
[51] DOI: 10.1080/01969727408546062 · Zbl 0304.68094 · doi:10.1080/01969727408546062
[52] DOI: 10.1109/91.873580 · doi:10.1109/91.873580
[53] DOI: 10.1109/72.914517 · doi:10.1109/72.914517
[54] DOI: 10.1109/42.996338 · doi:10.1109/42.996338
[55] DOI: 10.1016/S0167-8655(02)00401-4 · Zbl 1048.68078 · doi:10.1016/S0167-8655(02)00401-4
[56] DOI: 10.1016/j.patrec.2004.03.008 · doi:10.1016/j.patrec.2004.03.008
[57] DOI: 10.1016/j.patcog.2005.06.005 · Zbl 1079.68609 · doi:10.1016/j.patcog.2005.06.005
[58] Torres A., J. Biomed. Biotechnol. 6 pp 1–
[59] DOI: 10.1016/j.patcog.2006.07.011 · Zbl 1118.68133 · doi:10.1016/j.patcog.2006.07.011
[60] DOI: 10.1007/s00500-006-0111-5 · Zbl 05156927 · doi:10.1007/s00500-006-0111-5
[61] DOI: 10.1016/S0031-3203(97)00113-1 · Zbl 05467707 · doi:10.1016/S0031-3203(97)00113-1
[62] DOI: 10.1016/S0165-1684(98)00239-4 · Zbl 0923.68119 · doi:10.1016/S0165-1684(98)00239-4
[63] DOI: 10.1109/34.85677 · Zbl 05112080 · doi:10.1109/34.85677
[64] Littell R. C., SAS System for Mixed Models (2006)
[65] B. Muthén, Handbook of Quantitative Methodology for the Social Sciences, ed. D. Kaplan (Newbury Park, CA, Sage, 2004) pp. 345–368.
[66] DOI: 10.1542/peds.111.6.1318 · doi:10.1542/peds.111.6.1318
[67] DOI: 10.1111/j.1532-7078.2007.tb00244.x · doi:10.1111/j.1532-7078.2007.tb00244.x
[68] DOI: 10.1111/j.1467-8624.2006.00909.x · doi:10.1111/j.1467-8624.2006.00909.x
[69] DOI: 10.1080/00207160701286133 · Zbl 1114.92032 · doi:10.1080/00207160701286133
[70] DOI: 10.1385/NI:2:3:303 · doi:10.1385/NI:2:3:303
[71] DOI: 10.1142/S0219622008003204 · Zbl 05506004 · doi:10.1142/S0219622008003204
[72] DOI: 10.1142/S0219622006002271 · doi:10.1142/S0219622006002271
[73] DOI: 10.1142/S0219622007002368 · Zbl 1135.68385 · doi:10.1142/S0219622007002368
[74] DOI: 10.1142/S0219622006001885 · doi:10.1142/S0219622006001885
[75] DOI: 10.1142/S0219622006002283 · doi:10.1142/S0219622006002283
[76] L. N. de Castro and J. Timmis, Artificial Neural Networks in Pattern Recognition, eds. J. M. Corchado, L. Alonso and C. Fyfe (University of Paisley, UK, 2002) pp. 67–84.
[77] DOI: 10.1037/0022-006X.73.4.699 · doi:10.1037/0022-006X.73.4.699
[78] Hui L., Struct. Equat. Model. 14 pp 247–
[79] DOI: 10.2307/2530695 · Zbl 0625.62052 · doi:10.2307/2530695
[80] DOI: 10.1023/B:JOPA.0000008157.26321.3c · doi:10.1023/B:JOPA.0000008157.26321.3c
[81] DOI: 10.2333/bhmk.29.81 · Zbl 1017.62125 · doi:10.2333/bhmk.29.81
[82] DOI: 10.1093/biostatistics/3.4.459 · Zbl 1138.62365 · doi:10.1093/biostatistics/3.4.459
[83] DOI: 10.1037/1082-989X.4.2.139 · doi:10.1037/1082-989X.4.2.139
[84] DOI: 10.1198/016214501753168389 · Zbl 1017.62064 · doi:10.1198/016214501753168389
[85] Rubin D. B., J. Bus. Econ. Stat. 4 pp 87–
[86] DOI: 10.1002/9780470316696 · doi:10.1002/9780470316696
[87] DOI: 10.1080/01621459.2000.10473910 · doi:10.1080/01621459.2000.10473910
[88] DOI: 10.1080/00221325.1986.9914526 · doi:10.1080/00221325.1986.9914526
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.