# I have a dataset containing family information of married couples, which have around 10 variables & # 600+ observations.
# Independent variables are ~ gender, age, years married, children, religion etc.
# I have one response variable which is number of extra marital affairs.
# Now, I want to know what all factor influence the chances of extra marital affair.
# Since extra marital affair is a binary variable (either a person will have or not),
# so we can fit logistic regression model here to predict the probability of extra marital affair.
# install.packages('AER')
library('AER')
## Warning: package 'AER' was built under R version 3.5.1
## Loading required package: car
## Warning: package 'car' was built under R version 3.5.1
## Loading required package: carData
## Warning: package 'carData' was built under R version 3.5.1
## Loading required package: lmtest
## Warning: package 'lmtest' was built under R version 3.5.1
## Loading required package: zoo
## Warning: package 'zoo' was built under R version 3.5.1
##
## Attaching package: 'zoo'
## The following objects are masked from 'package:base':
##
## as.Date, as.Date.numeric
## Loading required package: sandwich
## Warning: package 'sandwich' was built under R version 3.5.1
## Loading required package: survival
library(plyr)
## Warning: package 'plyr' was built under R version 3.5.1
affairs <- data("Affairs")
View(Affairs)
affairs1 <- Affairs
summary(affairs1)
## affairs gender age yearsmarried children
## Min. : 0.000 female:315 Min. :17.50 Min. : 0.125 no :171
## 1st Qu.: 0.000 male :286 1st Qu.:27.00 1st Qu.: 4.000 yes:430
## Median : 0.000 Median :32.00 Median : 7.000
## Mean : 1.456 Mean :32.49 Mean : 8.178
## 3rd Qu.: 0.000 3rd Qu.:37.00 3rd Qu.:15.000
## Max. :12.000 Max. :57.00 Max. :15.000
## religiousness education occupation rating
## Min. :1.000 Min. : 9.00 Min. :1.000 Min. :1.000
## 1st Qu.:2.000 1st Qu.:14.00 1st Qu.:3.000 1st Qu.:3.000
## Median :3.000 Median :16.00 Median :5.000 Median :4.000
## Mean :3.116 Mean :16.17 Mean :4.195 Mean :3.932
## 3rd Qu.:4.000 3rd Qu.:18.00 3rd Qu.:6.000 3rd Qu.:5.000
## Max. :5.000 Max. :20.00 Max. :7.000 Max. :5.000
table(affairs1$affairs)
##
## 0 1 2 3 7 12
## 451 34 17 19 42 38
affairs1$ynaffairs[affairs1$affairs > 0] <- 1
affairs1$ynaffairs[affairs1$affairs == 0] <- 0
affairs1$gender <- as.factor(revalue(Affairs$gender,c("male"=1, "female"=0)))
affairs1$children <- as.factor(revalue(Affairs$children,c("yes"=1, "no"=0)))
# sum(is.na(claimants))
# claimants <- na.omit(claimants) # Omitting NA values from the Data
# na.omit => will omit the rows which has atleast 1 NA value
View(affairs1)
colnames(affairs1)
## [1] "affairs" "gender" "age" "yearsmarried"
## [5] "children" "religiousness" "education" "occupation"
## [9] "rating" "ynaffairs"
class(affairs1)
## [1] "data.frame"
attach(affairs1)
## The following object is masked _by_ .GlobalEnv:
##
## affairs
# Preparing a linear regression
mod_lm <- lm(ynaffairs ~ factor(gender) + age+ yearsmarried+ factor(children) + religiousness+
education+occupation+rating, data = affairs1)
summary(mod_lm)
##
## Call:
## lm(formula = ynaffairs ~ factor(gender) + age + yearsmarried +
## factor(children) + religiousness + education + occupation +
## rating, data = affairs1)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.6336 -0.2691 -0.1632 0.1151 1.0659
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.736107 0.151502 4.859 1.51e-06 ***
## factor(gender)1 0.045201 0.040022 1.129 0.259180
## age -0.007420 0.003013 -2.463 0.014057 *
## yearsmarried 0.015981 0.005491 2.911 0.003743 **
## factor(children)1 0.054487 0.046642 1.168 0.243198
## religiousness -0.053698 0.014881 -3.608 0.000334 ***
## education 0.003078 0.008542 0.360 0.718699
## occupation 0.005913 0.011838 0.499 0.617643
## rating -0.087455 0.015984 -5.472 6.59e-08 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.4122 on 592 degrees of freedom
## Multiple R-squared: 0.1066, Adjusted R-squared: 0.09452
## F-statistic: 8.829 on 8 and 592 DF, p-value: 1.884e-11
pred1 <- predict(mod_lm,affairs1)
pred1
## 4 5 11 16 23
## 0.2524486205 0.1136406616 0.4321912829 0.0376692455 0.3480920311
## 29 44 45 47 49
## 0.0598488898 0.2579352306 0.2620791697 0.4583221131 0.0567108409
## 50 55 64 80 86
## 0.6216067585 0.2256431201 0.1962503502 0.2215084253 0.1077280175
## 93 108 114 115 116
## 0.3466765722 0.4650551913 0.1527463202 0.1309746469 0.2543916193
## 123 127 129 134 137
## 0.1309746469 0.1309746469 0.2642586794 0.1635916625 0.2649915195
## 139 147 151 153 155
## 0.1371315087 0.0507222008 0.2425877369 0.1798161294 0.2251484823
## 162 163 165 168 170
## 0.3837409472 0.0772767840 0.1108171637 0.1730030818 0.3496346145
## 172 184 187 192 194
## 0.1984896163 0.2505913693 0.0666100566 0.2663790946 0.0416403963
## 210 217 220 224 227
## 0.1494311683 0.2496820346 0.0102214562 0.5502034627 0.2160492556
## 228 239 241 245 249
## 0.1624309141 0.1409295924 0.3464103939 0.0743402739 0.4036813253
## 262 265 267 269 271
## -0.0600869029 0.3500368589 0.0906467715 0.3276864727 0.2759835016
## 277 290 292 293 295
## 0.5143437930 0.1511124967 0.1619324074 0.0650247943 0.2905843787
## 299 320 321 324 334
## 0.3226369710 0.3071801475 0.1189888355 0.1792709007 0.1881538240
## 351 355 361 362 366
## 0.2892527528 0.2534822847 0.1709893027 -0.0301189416 0.1872405445
## 370 374 378 381 382
## 0.0832058255 0.0739338692 0.0966869358 0.3163429796 0.4197447991
## 383 384 400 403 409
## 0.1740615120 0.1317640934 0.1888072044 0.0772767840 0.4547277970
## 412 413 416 418 422
## 0.2224269094 0.0975239701 0.1946188949 0.1264885131 0.0421019306
## 435 439 445 447 448
## 0.2638114206 0.2226850778 0.3086975944 0.1019566552 0.2938503251
## 449 478 482 486 489
## 0.2890798445 0.4239501283 0.2352559170 0.2128241583 0.2510943890
## 490 491 492 503 508
## 0.0618971579 0.4887294995 0.4022621495 0.2844822722 0.2459323462
## 509 512 515 517 532
## 0.0932980309 0.3547211485 -0.0166214340 0.0438261073 0.2820294063
## 533 535 537 538 543
## 0.2504314904 0.2676737214 0.1790666736 0.4931724126 0.2122731326
## 547 550 558 571 578
## 0.0711858470 0.1914956279 -0.0981662678 0.1685949719 0.2155259875
## 583 586 594 597 602
## 0.1914450369 0.4720898711 0.1762423448 -0.0120574665 0.1862637781
## 603 604 612 613 621
## 0.2185178798 0.1853168054 0.3685930104 0.4749712783 0.0772767840
## 627 630 631 632 639
## 0.1922344834 0.0969311535 0.0316782528 0.1140413903 0.1480650534
## 645 647 648 651 655
## 0.3280634506 0.2026733904 0.0997851969 0.3660068839 0.3621529339
## 667 670 671 673 701
## 0.3669167191 0.1310583414 0.3537050497 0.1384319902 0.3324936169
## 705 706 709 717 719
## 0.3356901051 0.2654336375 0.2829606207 0.3174076314 0.3184658829
## 723 724 726 734 735
## 0.3066198764 0.2074167523 0.1564468032 0.1274641685 0.4226104480
## 736 737 739 743 745
## 0.2929943868 0.3270314503 0.3774794551 0.4208430412 0.3815187596
## 747 751 752 754 760
## 0.3837409472 0.3467496527 0.4150631573 0.3428153453 0.1964118982
## 763 774 776 779 784
## 0.2003820726 0.1944694285 0.1860776216 0.1297441260 0.2602823028
## 788 794 795 798 800
## 0.2827580337 0.4100148069 0.1048774070 0.3137749448 0.3023766965
## 803 807 812 820 823
## 0.0975239701 -0.0104645722 0.1098937997 0.2927125310 0.1059420587
## 830 843 848 851 854
## 0.4030081107 0.1675827421 0.3480038287 0.2430181112 0.2638030386
## 856 857 859 863 865
## 0.3630617137 0.0488192504 0.2755017358 0.0063254509 0.0964744318
## 867 870 873 875 876
## 0.0990526483 0.0733903043 0.2458277159 0.1789900194 0.4197447991
## 877 880 903 904 905
## 0.1076581851 0.1957432885 0.1686271217 0.3274750496 0.4350035354
## 908 909 910 912 914
## 0.4179261584 0.4056951044 0.1189051410 0.3785215213 0.0266573521
## 915 916 920 921 925
## 0.0173361598 0.1784297483 0.0831894281 0.2963304358 0.5295670869
## 926 929 931 945 947
## 0.2598758675 0.4089062955 0.1548650715 0.2247365400 0.3635721098
## 949 950 961 965 966
## 0.3498720050 0.2222856388 0.3765391779 0.1301665980 0.1309910442
## 967 987 990 992 995
## -0.1150995243 0.0777714218 0.2867221773 0.3604110092 0.1804091952
## 1009 1021 1026 1027 1030
## 0.3371203458 0.1995949595 0.1113787958 0.1562356943 0.2448901483
## 1031 1034 1037 1038 1039
## 0.3090736189 0.2302350163 0.2632787030 0.2215491167 0.0564403415
## 1045 1046 1054 1059 1063
## 0.2101051758 0.0948364234 0.1574719648 0.1711558107 0.0688640413
## 1068 1070 1072 1073 1077
## 0.2948720999 -0.0595984675 -0.0040486083 0.2324436645 0.2640472563
## 1081 1083 1084 1086 1087
## 0.3441242148 0.2690619547 0.4259016609 0.1859729912 0.2829229826
## 1089 1096 1102 1103 1107
## 0.2285402610 0.2521874791 0.1274203089 0.2123514812 0.2796898309
## 1109 1115 1119 1124 1126
## 0.2531594041 0.2923750433 0.1400378489 0.1957432885 0.2419434411
## 1128 1129 1130 1133 1140
## 0.2139725527 0.1011672087 0.4036813253 0.1274660014 0.2187620975
## 1143 1146 1153 1156 1157
## 0.1419093597 0.3964598117 0.3975778599 0.1248177851 0.2707104884
## 1158 1160 1161 1166 1177
## 0.2176385603 0.6099142305 0.2409401017 0.1970049640 0.1348589216
## 1178 1180 1187 1191 1195
## 0.2459323462 0.0915329774 0.2898239985 0.4854354519 0.1200417411
## 1207 1208 1209 1211 1215
## 0.5737634786 0.2222985108 0.2929787094 0.2046961964 0.1989287031
## 1221 1226 1229 1231 1234
## 0.3333876937 0.2895682799 0.1822917709 0.3442035786 0.2463627205
## 1235 1242 1245 1260 1266
## 0.3418295141 0.1106529214 0.4362695279 0.4089596919 0.3835124879
## 1271 1273 1276 1280 1282
## 0.1837396029 0.1761760724 0.2737771868 0.1984896163 0.1160506615
## 1285 1295 1298 1299 1304
## 0.3570158089 0.0001145478 0.1399821192 0.4087735353 0.2056970660
## 1305 1311 1314 1319 1322
## 0.4865523488 0.1964118982 0.2897798899 0.1227224273 0.1874921326
## 1324 1327 1328 1330 1332
## 0.2459929295 0.1913037430 0.0406196041 0.4056951044 0.0761572202
## 1333 1336 1341 1344 1352
## 0.3474917382 0.1361547423 0.0995597099 0.3687545585 0.2739099470
## 1358 1359 1361 1364 1368
## 0.0209688464 0.1214030345 0.3169191960 0.1947794180 0.1738033881
## 1384 1390 1393 1394 1402
## 0.0393971270 0.1309746469 0.2317999819 0.1897992087 0.0820472647
## 1407 1408 1412 1413 1416
## 0.2138488195 0.3870855565 0.1985222983 0.0798465132 0.1930289310
## 1417 1418 1419 1420 1423
## 0.1896714497 0.1973491472 0.2019125742 0.0871411691 0.1739034853
## 1424 1432 1433 1437 1438
## 0.4266442278 -0.0255089841 0.1588539765 0.2496539334 0.2605484812
## 1439 1446 1450 1451 1452
## 0.2451962961 0.4350035354 0.2781824221 0.1436687067 0.2267809141
## 1453 1456 1464 1469 1473
## -0.0051555942 0.0982350680 0.4085827140 0.2163015278 0.2526601562
## 1481 1482 1496 1497 1504
## 0.0938725529 0.0432531209 0.0746512000 0.0572662647 0.0449525411
## 1513 1515 1534 1535 1536
## 0.2095392287 0.4420583430 0.1701509556 0.1501625271 0.0816284958
## 1540 1551 1555 1557 1566
## 0.2568268940 -0.0594959591 0.1122504114 0.1635916625 0.0823760333
## 1567 1576 1584 1585 1590
## 0.1980424971 0.0711858470 0.0789485675 0.3707564593 0.1748773979
## 1594 1595 1603 1608 1609
## 0.3732439791 0.2941226440 0.2505120055 0.1408153768 0.0626234850
## 1615 1616 1617 1620 1621
## 0.1593647163 0.1218052789 0.2494535752 0.0416403963 0.1793050944
## 1637 1638 1650 1654 1665
## 0.2420104295 0.1163996965 0.2031290312 0.2468697821 0.4688144165
## 1670 1671 1675 1688 1691
## 0.0115094153 0.3345523110 0.2064441830 0.1309746469 0.2986251751
## 1695 1698 1704 1705 1711
## 0.0630927177 0.2721465879 0.3563169441 0.5187372958 0.4317123225
## 1719 1723 1726 1749 1752
## 0.0368088208 0.1905851538 0.1606171924 0.1836423469 0.3572436293
## 1754 1758 1761 1773 1775
## 0.0577609024 0.0449525411 0.2305182310 0.3010215913 0.3412573788
## 1786 1793 1799 1803 1806
## 0.0515268765 0.3143292005 0.1624928655 0.2723022812 0.1227032247
## 1807 1808 1814 1815 1818
## 0.3278558965 0.2254794174 0.2762585301 0.2102233289 0.0474693459
## 1827 1834 1835 1843 1846
## 0.3705179176 -0.0021647647 0.3905471676 0.1887148111 0.1807404953
## 1850 1851 1854 1859 1861
## 0.1632203606 0.2521173379 0.6335648069 0.0275289678 0.3812263088
## 1866 1873 1875 1885 1892
## 0.1199614048 0.1960746596 0.1865079958 0.3408194792 0.2785148728
## 1895 1896 1897 1899 1904
## 0.3964755701 0.0803552149 0.2107394010 -0.0123070301 0.4626575547
## 1905 1908 1916 1918 1920
## 0.2340489968 0.3104460939 0.2267009446 0.3309900571 0.5620496480
## 1930 1940 1947 1949 1951
## 0.3498720050 0.1679908413 0.2402058586 0.1189051410 0.1385647504
## 1952 1960 9001 9012 9023
## 0.0573747639 0.1250620028 0.1803129669 0.5296318436 0.3719376197
## 9029 6 12 43 53
## 0.2521173379 0.1730756809 0.1140425584 0.4484436642 0.4588435742
## 67 79 122 126 133
## 0.0468063593 0.1556545180 0.4839151593 0.2004478444 0.4346265575
## 138 154 159 174 176
## 0.4578336777 0.4328010896 0.2659835238 0.4140643258 0.5357887054
## 181 182 186 189 204
## 0.5058631142 0.2447940517 0.4480891401 0.0832058255 0.1906038845
## 215 232 233 252 253
## 0.0425329962 0.2282269132 0.1328101330 0.2303972034 0.3909819113
## 274 275 287 288 325
## 0.2328578064 0.3512255915 0.4075968828 0.1402105785 0.3276612062
## 328 344 353 354 367
## 0.1973491472 0.5138621848 -0.0207823742 0.1222075233 0.2956001406
## 369 390 392 423 432
## 0.4407834193 0.3964298417 0.2470969635 0.3294771373 0.4454424424
## 436 483 513 516 518
## 0.4169324849 0.2934733472 0.1805458243 0.6095906490 0.1859729912
## 520 526 528 553 576
## 0.1624928655 0.4592595491 0.1948020177 0.0920022102 0.1487701551
## 611 625 635 646 657
## 0.1872592752 0.4348707753 0.3670625088 0.1669571770 0.1687605268
## 659 666 679 729 755
## 0.2822573652 0.2406081031 0.2837810547 0.5944207515 0.4742578279
## 758 770 786 797 811
## 0.3164430769 0.2962855997 0.2888019747 0.3660355143 0.3687545585
## 834 858 885 893 927
## 0.2007641892 0.3250123252 0.4016116397 0.4899617148 0.6031142124
## 928 933 951 968 972
## 0.2002819753 0.3605470752 0.4884948186 0.0848382572 0.3272842283
## 975 977 981 986 1002
## 0.3497550032 0.4276430593 0.1848469337 0.3234449812 0.2200197472
## 1007 1011 1035 1050 1056
## 0.3779610633 0.3198990120 0.4061447298 0.3632232617 0.3619415109
## 1057 1075 1080 1125 1131
## 0.1865332623 0.5357887054 0.4468822200 0.3552618291 0.3676842307
## 1138 1150 1163 1169 1198
## 0.3458144245 0.3669167191 0.2743160603 0.1551857857 0.5079495407
## 1204 1218 1230 1236 1247
## 0.1931222010 0.5717170214 0.4791999272 0.3440708184 0.5004213401
## 1259 1294 1353 1370 1427
## 0.5779263072 0.0320008192 0.2591835057 0.4366999546 0.4653056113
## 1445 1460 1480 1505 1543
## 0.4394525892 0.4350821983 0.4225570516 0.6369094161 0.4225022963
## 1548 1550 1561 1564 1573
## 0.5836842304 0.0969509838 0.4811405640 0.4256354825 0.1276683956
## 1575 1599 1622 1629 1664
## 0.1469958937 0.3229769166 -0.0659362673 0.2301851413 0.3863467011
## 1669 1674 1682 1685 1697
## 0.5285976530 0.1827290176 0.2595209934 0.2127999260 0.1885410260
## 1716 1730 1731 1732 1743
## 0.4002869978 0.3433319630 0.2185021215 0.4890383569 0.4251250864
## 1751 1757 1763 1766 1772
## 0.3808240644 0.4267556853 0.3870855565 0.3269477557 0.4233064457
## 1776 1782 1784 1791 1831
## 0.2293365774 0.2384487211 0.2721775755 0.3011493503 0.5176884023
## 1840 1844 1856 1876 1929
## 0.3675755527 0.5187372958 0.4971820771 0.3719910161 0.3477438526
## 1935 1938 1941 1954 1959
## 0.4636530518 0.2543095887 0.3407091728 0.3012894641 0.5392970174
## 9010
## 0.2434970716
# plot(affairs,pred1)
# We can no way use the linear regression technique to classify the data
plot(pred1)
# We can also include NA values but where ever it finds NA value
# probability values obtained using the glm will also be NA
# So they can be either filled using imputation technique or
# exlclude those values
# GLM function use sigmoid curve to produce desirable results
# The output of sigmoid function lies in between 0-1
model <- glm(ynaffairs ~ factor(gender) + age+ yearsmarried+ factor(children) + religiousness+
education+occupation+rating, data = affairs1,family = "binomial")
# To calculate the odds ratio manually we going r going to take exp of coef(model)
exp(coef(model))
## (Intercept) factor(gender)1 age yearsmarried
## 3.9640180 1.3235091 0.9567099 1.0994093
## factor(children)1 religiousness education occupation
## 1.4883560 0.7227292 1.0212740 1.0314027
## rating
## 0.6259691
# Confusion matrix table
prob <- predict(model,affairs1,type="response")
summary(model)
##
## Call:
## glm(formula = ynaffairs ~ factor(gender) + age + yearsmarried +
## factor(children) + religiousness + education + occupation +
## rating, family = "binomial", data = affairs1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.5713 -0.7499 -0.5690 -0.2539 2.5191
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.37726 0.88776 1.551 0.120807
## factor(gender)1 0.28029 0.23909 1.172 0.241083
## age -0.04426 0.01825 -2.425 0.015301 *
## yearsmarried 0.09477 0.03221 2.942 0.003262 **
## factor(children)1 0.39767 0.29151 1.364 0.172508
## religiousness -0.32472 0.08975 -3.618 0.000297 ***
## education 0.02105 0.05051 0.417 0.676851
## occupation 0.03092 0.07178 0.431 0.666630
## rating -0.46845 0.09091 -5.153 2.56e-07 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 675.38 on 600 degrees of freedom
## Residual deviance: 609.51 on 592 degrees of freedom
## AIC: 627.51
##
## Number of Fisher Scoring iterations: 4
# We are going to use NULL and Residual Deviance to compare the between different models
# Confusion matrix and considering the threshold value as 0.5
confusion<-table(prob>0.5,affairs1$ynaffairs)
confusion
##
## 0 1
## FALSE 435 125
## TRUE 16 25
# Model Accuracy
Accuracy<-sum(diag(confusion)/sum(confusion))
Accuracy # 76.53
## [1] 0.765391
# Creating empty vectors to store predicted classes based on threshold value
pred_values <- NULL
yes_no <- NULL
pred_values <- ifelse(prob>=0.5,1,0)
yes_no <- ifelse(prob>=0.5,"yes","no")
# Creating new column to store the above values
affairs1[,"prob"] <- prob
affairs1[,"pred_values"] <- pred_values
affairs1[,"yes_no"] <- yes_no
View(affairs1[,c(1,9:11)])
table(affairs1$ynaffairs,affairs1$pred_values)
##
## 0 1
## 0 435 16
## 1 125 25
# Calculate the below metrics
# precision | recall | True Positive Rate | False Positive Rate | Specificity | Sensitivity
# from the above table - 59
# ROC Curve => used to evaluate the betterness of the logistic model
# more area under ROC curve better is the model
# We will use ROC curve for any classification technique not only for logistic
# install.packages("ROCR")
library(ROCR)
## Warning: package 'ROCR' was built under R version 3.5.1
## Loading required package: gplots
## Warning: package 'gplots' was built under R version 3.5.1
##
## Attaching package: 'gplots'
## The following object is masked from 'package:stats':
##
## lowess
rocrpred<-prediction(prob,affairs1$ynaffairs)
rocrperf<-performance(rocrpred,'tpr','fpr')
str(rocrperf)
## Formal class 'performance' [package "ROCR"] with 6 slots
## ..@ x.name : chr "False positive rate"
## ..@ y.name : chr "True positive rate"
## ..@ alpha.name : chr "Cutoff"
## ..@ x.values :List of 1
## .. ..$ : num [1:570] 0 0 0.00222 0.00443 0.00443 ...
## ..@ y.values :List of 1
## .. ..$ : num [1:570] 0 0.00667 0.00667 0.00667 0.01333 ...
## ..@ alpha.values:List of 1
## .. ..$ : num [1:570] Inf 0.725 0.709 0.709 0.696 ...
plot(rocrperf,colorize=T,text.adj=c(-0.2,1.7))
# More area under the ROC Curve better is the logistic regression model obtained
## Getting cutt off or threshold value along with true positive and false positive rates in a data frame
str(rocrperf)
## Formal class 'performance' [package "ROCR"] with 6 slots
## ..@ x.name : chr "False positive rate"
## ..@ y.name : chr "True positive rate"
## ..@ alpha.name : chr "Cutoff"
## ..@ x.values :List of 1
## .. ..$ : num [1:570] 0 0 0.00222 0.00443 0.00443 ...
## ..@ y.values :List of 1
## .. ..$ : num [1:570] 0 0.00667 0.00667 0.00667 0.01333 ...
## ..@ alpha.values:List of 1
## .. ..$ : num [1:570] Inf 0.725 0.709 0.709 0.696 ...
rocr_cutoff <- data.frame(cut_off = rocrperf@alpha.values[[1]],fpr=rocrperf@x.values,tpr=rocrperf@y.values)
colnames(rocr_cutoff) <- c("cut_off","FPR","TPR")
View(rocr_cutoff)
library(dplyr)
## Warning: package 'dplyr' was built under R version 3.5.1
##
## Attaching package: 'dplyr'
## The following objects are masked from 'package:plyr':
##
## arrange, count, desc, failwith, id, mutate, rename, summarise,
## summarize
## The following object is masked from 'package:car':
##
## recode
## The following objects are masked from 'package:stats':
##
## filter, lag
## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, union
rocr_cutoff$cut_off <- round(rocr_cutoff$cut_off,6)
# Sorting data frame with respect to tpr in decreasing order
rocr_cutoff <- arrange(rocr_cutoff,desc(TPR))
## Warning: package 'bindrcpp' was built under R version 3.5.1
View(rocr_cutoff)