Correcting Classifiers for Sample Selection Bias in Two-Phase Case-Control Studies

<div>Normal quantile-quantile plots for main features <svg height="12.0784pt" id="M253" style="vertical-align:-0.04979992pt" version="1.1" viewbox="-0.0498162 -12.0286 9.23375 12.0784" width="9.23375pt" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink"><g transform="matrix(.013,0,0,-0.013,.274,-3.001)"><path d="M658 624L641 649C599 612 554 595 493 595C444 595 402 603 342 615C268 634 232 641 184 641C129 641 57 621 0 557L17 532C67 566 113 586 176 586C230 586 265 577 316 565C380 550 430 540 482 540C540 540 611 558 658 624Z" id="g119-113"></path></g><g transform="matrix(.013,0,0,-0.013,0,0)"><path d="M700 0V35C631 42 609 53 567 118C509 208 460 288 401 381C449 447 492 506 526 546C567 595 594 607 658 615V650H417V615C481 603 484 593 465 557C441 512 406 457 375 417C333 483 302 536 285 568C273 591 271 603 326 615V650H25V615C93 608 120 591 148 544C201 461 242 391 290 312C233 226 163 135 127 90C98 57 63 41 8 35V0H259V35C192 46 183 54 207 97C228 137 269 202 316 273C390 149 413 111 423 93C440 61 445 47 379 35V0H700Z" id="g13-86"></path></g></svg> in real data set. For visualization purposes, we only displayed a random sample of 10,000 observations instead of the full data set of size <span class="nowrap"><svg height="11.7978pt" id="M254" style="vertical-align:-0.2063999pt" version="1.1" viewbox="-0.0498162 -11.5914 17.5728 11.7978" width="17.5728pt" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink"><g transform="matrix(.013,0,0,-0.013,0,0)"><path d="M384 0V27C293 34 287 42 287 114V635C232 613 172 594 109 583V559L157 557C201 555 205 550 205 499V114C205 42 199 34 109 27V0H384Z" id="g113-50"></path></g><g transform="matrix(.013,0,0,-0.013,6.24,0)"><path d="M241 635C89 635 35 457 35 312C35 153 89 -12 240 -12C390 -12 443 166 443 312C443 466 390 635 241 635ZM238 602C329 602 354 454 354 312C354 172 330 22 240 22C152 22 124 173 124 313S148 602 238 602Z" id="g113-49"></path></g><g transform="matrix(.0091,0,0,-0.0091,12.48,-5.741)"><path d="M141 347C171 480 264 541 327 570C364 587 403 599 436 605L429 641C389 634 339 622 300 609C191 570 37 458 37 239C37 86 126 -12 245 -12C367 -12 454 89 454 210C454 314 380 397 272 397C252 397 230 390 208 380L141 347ZM231 338C323 338 366 257 366 174C366 108 340 27 263 27C180 27 129 123 129 244C129 268 130 292 135 310C158 323 193 338 231 338Z" id="g50-55"></path></g></svg>.</span></div>

Computational and Mathematical Methods in Medicine

fig7

Figure 7

Figure 7: Correcting Classifiers for Sample Selection Bias in Two-Phase Case-Control Studies 