%PDF-1.5
%
1 0 obj
<>/Metadata 824 0 R/Outlines 56 0 R/Pages 2 0 R/StructTreeRoot 66 0 R/Type/Catalog>>
endobj
824 0 obj
<>stream
Longitudinal data often includes time-dependent covariates, which must be accounted for with an appropriate model. Although a number of models have been proposed to analyze time-dependent covariates, most approaches constrain the effect of each covariate on the outcome to be constant across time. Irimata, Broatch, and Wilson (2017) introduced a partitioned generalized method of moments (GMM) model that used only valid moment conditions to estimate the differing relationships within longitudinal data. This model provides insight into potential lagged effects of a given covariate on the response in a later time period. Each regression coefficient is estimated using moment conditions corresponding to the respective time period. Irimata and Wilson (2017) presented a SAS® macro for fitting this partitioned GMM model for binary outcomes using SAS/IML® software. We extended the %partitionedGMM macro to allow for either continuous or binary outcomes. In this paper, we also expanded this macro to fit time-independent covariates. The performance and use of this macro are demonstrated through the analysis of two examplesone with a continuous outcome and one with a binary outcome. <br/><br/>Kyle Irimata, Arizona State University <br/><br/>Jeffrey Wilson, Arizona State University
Session 2661
en
jeff.foxx@sas.com
Longitudinal data often includes time-dependent covariates, which must be accounted for with an appropriate model. Although a number of models have been proposed to analyze time-dependent covariates, most approaches constrain the effect of each covariate on the outcome to be constant across time. Irimata, Broatch, and Wilson (2017) introduced a partitioned generalized method of moments (GMM) model that used only valid moment conditions to estimate the differing relationships within longitudinal data. This model provides insight into potential lagged effects of a given covariate on the response in a later time period. Each regression coefficient is estimated using moment conditions corresponding to the respective time period. Irimata and Wilson (2017) presented a SAS® macro for fitting this partitioned GMM model for binary outcomes using SAS/IML® software. We extended the %partitionedGMM macro to allow for either continuous or binary outcomes. In this paper, we also expanded this macro to fit time-independent covariates. The performance and use of this macro are demonstrated through the analysis of two examplesone with a continuous outcome and one with a binary outcome. <br/><br/>Kyle Irimata, Arizona State University <br/><br/>Jeffrey Wilson, Arizona State University
2018-04-03T15:46:57.667-04:00
ff337dbf9c78415f26ee68024892229e89c519f1
560403
application/pdf
2018-04-03T15:46:57.266-04:00
Using SAS® to Estimate Lagged Coefficients with the %partitionedGMM Macro
thirdparty
support:sgf-papers
year:2018
industry:3150
software:IML
support:sgf-papers/session-type/breakout
support:sgf-papers/skill-level/intermediate
support:sgf-papers/topic/analytics/statistics
support:customer-roles/statistician
endstream
endobj
56 0 obj
<>
endobj
2 0 obj
<>
endobj
66 0 obj
<>
endobj
68 0 obj
<>
endobj
67 0 obj
<>
endobj
71 0 obj
[70 0 R 73 0 R 74 0 R 75 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 76 0 R 77 0 R 78 0 R 79 0 R 80 0 R 81 0 R]
endobj
83 0 obj
[82 0 R 84 0 R 85 0 R 86 0 R 87 0 R 87 0 R 87 0 R 87 0 R 87 0 R 87 0 R 87 0 R 87 0 R 88 0 R 88 0 R 88 0 R 88 0 R 89 0 R 89 0 R 89 0 R 89 0 R 90 0 R 90 0 R 90 0 R 90 0 R 91 0 R 91 0 R 91 0 R 92 0 R 92 0 R 92 0 R 92 0 R 93 0 R 93 0 R 93 0 R 93 0 R 93 0 R 93 0 R 94 0 R 94 0 R 94 0 R 94 0 R 94 0 R 94 0 R 95 0 R 95 0 R 95 0 R 95 0 R 96 0 R 96 0 R 96 0 R 96 0 R 97 0 R 97 0 R 97 0 R 97 0 R 98 0 R 99 0 R]
endobj
102 0 obj
[101 0 R 105 0 R 106 0 R 107 0 R 107 0 R 107 0 R 107 0 R 107 0 R 107 0 R 107 0 R 107 0 R 107 0 R 107 0 R 108 0 R 108 0 R 108 0 R 108 0 R 109 0 R 109 0 R 109 0 R 109 0 R 110 0 R 110 0 R 110 0 R 110 0 R 111 0 R 111 0 R 111 0 R 111 0 R 112 0 R 112 0 R 112 0 R 112 0 R 113 0 R 113 0 R 113 0 R 113 0 R 113 0 R 113 0 R 114 0 R 114 0 R 114 0 R 114 0 R 115 0 R 115 0 R 115 0 R 115 0 R 116 0 R 116 0 R 116 0 R 116 0 R 117 0 R 117 0 R 117 0 R 117 0 R 118 0 R 119 0 R 121 0 R 122 0 R 123 0 R 124 0 R 125 0 R 120 0 R]
endobj
103 0 obj
<>
endobj
127 0 obj
[126 0 R 128 0 R 128 0 R 128 0 R 128 0 R 128 0 R 128 0 R 128 0 R 128 0 R 128 0 R 128 0 R 129 0 R 129 0 R 129 0 R 129 0 R 130 0 R 130 0 R 130 0 R 130 0 R 131 0 R 131 0 R 131 0 R 131 0 R 132 0 R 132 0 R 132 0 R 132 0 R 132 0 R 133 0 R 133 0 R 133 0 R 133 0 R 133 0 R 134 0 R 134 0 R 134 0 R 134 0 R 134 0 R 134 0 R 135 0 R 135 0 R 135 0 R 135 0 R 135 0 R 136 0 R 136 0 R 136 0 R 136 0 R 137 0 R 137 0 R 137 0 R 137 0 R 138 0 R 138 0 R 138 0 R 138 0 R 140 0 R 143 0 R 144 0 R 145 0 R 147 0 R 148 0 R 149 0 R 149 0 R 149 0 R 149 0 R 149 0 R 149 0 R 149 0 R 149 0 R 149 0 R 149 0 R 150 0 R 150 0 R 150 0 R 150 0 R 150 0 R 150 0 R 151 0 R 151 0 R 151 0 R 151 0 R 152 0 R 152 0 R 152 0 R 152 0 R 153 0 R 153 0 R 153 0 R 153 0 R 153 0 R 154 0 R 154 0 R 154 0 R 154 0 R 154 0 R 155 0 R 155 0 R 155 0 R 155 0 R 155 0 R 155 0 R 156 0 R 156 0 R 156 0 R 156 0 R 156 0 R 157 0 R 157 0 R 157 0 R 157 0 R 158 0 R 158 0 R 158 0 R 158 0 R 159 0 R 159 0 R 159 0 R 159 0 R 146 0 R]
endobj
141 0 obj
<>
endobj
162 0 obj
[161 0 R 165 0 R 166 0 R 167 0 R 169 0 R 171 0 R 174 0 R 175 0 R 179 0 R 181 0 R 183 0 R 185 0 R 187 0 R 188 0 R 191 0 R 193 0 R 195 0 R 197 0 R 199 0 R 201 0 R 203 0 R 205 0 R 206 0 R 209 0 R 211 0 R 213 0 R 215 0 R 217 0 R 219 0 R 221 0 R 223 0 R 224 0 R 227 0 R 229 0 R 231 0 R 233 0 R 235 0 R 237 0 R 239 0 R 241 0 R 242 0 R 245 0 R 247 0 R 249 0 R 251 0 R 253 0 R 255 0 R 257 0 R 259 0 R 260 0 R 264 0 R 266 0 R 268 0 R 270 0 R 272 0 R 274 0 R 276 0 R 277 0 R 281 0 R 283 0 R 285 0 R 287 0 R 289 0 R 291 0 R 293 0 R 294 0 R 298 0 R 300 0 R 302 0 R 304 0 R 306 0 R 308 0 R 310 0 R 311 0 R 314 0 R 316 0 R 318 0 R 320 0 R 322 0 R 324 0 R 326 0 R 328 0 R 329 0 R 333 0 R 335 0 R 337 0 R 339 0 R 341 0 R 343 0 R 345 0 R 346 0 R 350 0 R 352 0 R 354 0 R 356 0 R 358 0 R 360 0 R 362 0 R 363 0 R 367 0 R 369 0 R 371 0 R 373 0 R 375 0 R 377 0 R 379 0 R 380 0 R 383 0 R 385 0 R 387 0 R 389 0 R 391 0 R 393 0 R 395 0 R 397 0 R 398 0 R 402 0 R 404 0 R 406 0 R 408 0 R 410 0 R 412 0 R 414 0 R 415 0 R 419 0 R 421 0 R 423 0 R 425 0 R 427 0 R 429 0 R 431 0 R 432 0 R 436 0 R 438 0 R 440 0 R 442 0 R 444 0 R 446 0 R 448 0 R 449 0 R 452 0 R 454 0 R 456 0 R 458 0 R 460 0 R 462 0 R 464 0 R 466 0 R 467 0 R 471 0 R 473 0 R 475 0 R 477 0 R 479 0 R 481 0 R 483 0 R 484 0 R 488 0 R 490 0 R 492 0 R 494 0 R 496 0 R 498 0 R 500 0 R 501 0 R 505 0 R 507 0 R 509 0 R 511 0 R 513 0 R 515 0 R 517 0 R 518 0 R 519 0 R 168 0 R]
endobj
163 0 obj
<>
endobj
172 0 obj
<>
endobj
521 0 obj
[520 0 R 522 0 R 523 0 R 523 0 R 523 0 R 523 0 R 523 0 R 523 0 R 523 0 R 523 0 R 523 0 R 523 0 R 524 0 R 524 0 R 524 0 R 524 0 R 524 0 R 525 0 R 525 0 R 525 0 R 525 0 R 526 0 R 526 0 R 526 0 R 526 0 R 527 0 R 527 0 R 527 0 R 527 0 R 528 0 R 528 0 R 528 0 R 528 0 R 529 0 R 529 0 R 529 0 R 529 0 R 529 0 R 529 0 R 530 0 R 530 0 R 530 0 R 530 0 R 531 0 R 531 0 R 531 0 R 531 0 R 531 0 R 531 0 R 532 0 R 532 0 R 532 0 R 532 0 R 533 0 R 533 0 R 533 0 R 533 0 R 535 0 R 538 0 R 539 0 R 540 0 R 542 0 R 543 0 R 544 0 R 544 0 R 544 0 R 544 0 R 544 0 R 544 0 R 544 0 R 544 0 R 544 0 R 544 0 R 545 0 R 545 0 R 545 0 R 545 0 R 546 0 R 546 0 R 546 0 R 546 0 R 547 0 R 547 0 R 547 0 R 547 0 R 548 0 R 548 0 R 548 0 R 548 0 R 549 0 R 549 0 R 549 0 R 549 0 R 550 0 R 550 0 R 550 0 R 550 0 R 550 0 R 550 0 R 551 0 R 551 0 R 551 0 R 551 0 R 552 0 R 552 0 R 552 0 R 552 0 R 553 0 R 553 0 R 553 0 R 553 0 R 554 0 R 554 0 R 554 0 R 554 0 R 556 0 R 559 0 R 560 0 R 541 0 R]
endobj
536 0 obj
<>
endobj
557 0 obj
<>
endobj
562 0 obj
[561 0 R 564 0 R 566 0 R 569 0 R 570 0 R 574 0 R 576 0 R 578 0 R 580 0 R 583 0 R 584 0 R 587 0 R 589 0 R 591 0 R 593 0 R 595 0 R 597 0 R 599 0 R 601 0 R 602 0 R 605 0 R 607 0 R 609 0 R 611 0 R 613 0 R 615 0 R 617 0 R 619 0 R 620 0 R 623 0 R 625 0 R 627 0 R 629 0 R 631 0 R 633 0 R 635 0 R 637 0 R 638 0 R 641 0 R 643 0 R 645 0 R 647 0 R 649 0 R 651 0 R 653 0 R 655 0 R 656 0 R 660 0 R 662 0 R 664 0 R 666 0 R 668 0 R 670 0 R 672 0 R 673 0 R 676 0 R 678 0 R 680 0 R 682 0 R 684 0 R 686 0 R 688 0 R 690 0 R 691 0 R 695 0 R 697 0 R 699 0 R 701 0 R 703 0 R 705 0 R 707 0 R 708 0 R 711 0 R 713 0 R 715 0 R 717 0 R 719 0 R 721 0 R 723 0 R 725 0 R 726 0 R 730 0 R 732 0 R 734 0 R 736 0 R 738 0 R 740 0 R 742 0 R 743 0 R 746 0 R 748 0 R 750 0 R 752 0 R 754 0 R 756 0 R 758 0 R 760 0 R 761 0 R 765 0 R 767 0 R 769 0 R 771 0 R 773 0 R 775 0 R 777 0 R 778 0 R 779 0 R 780 0 R 781 0 R 782 0 R 563 0 R]
endobj
567 0 obj
<>
endobj
784 0 obj
[783 0 R 785 0 R 787 0 R 790 0 R 791 0 R 792 0 R 793 0 R 794 0 R 795 0 R 796 0 R 797 0 R 798 0 R 799 0 R 800 0 R 801 0 R 802 0 R 803 0 R 804 0 R 805 0 R 806 0 R]
endobj
788 0 obj
<>
endobj
786 0 obj
<>
endobj
50 0 obj
<>/MediaBox[0 0 612 792]/Parent 2 0 R/Resources<>/Font<>/ProcSet[/PDF/Text/ImageB/ImageC/ImageI]>>/StructParents 14/Tabs/S/Type/Page>>
endobj
51 0 obj
<>stream
xZRH}'iÔUX.
MY`vcamO?
*OGq?<.|}x(d6ϫY9ޮ*|tVUgq|z"gJdiO )x2w/6=OLW;QR%(hc{D5atrq*ud[%P$ʓI],ENdΣxpT漹*uĥ{:M`9
>o/M93v#GRqZSMh_g *2Ł`RRG%nN?g0{ͫ
Gc(+Qԩ{#.0n!_Rc="%N9O5F^ϖiY&}_+\Q=xegchS
X:7,by'^TPVMGWµ}9lzҋ$#LI(P:=D-mw;P!ui/Ğ
PB]'h8O~&<z2
XNސ|