R code

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609

getwd() # setwd("C:/Users/aseandi/My Library/COURSERA/Passion-Driven Statistics/datasets") getwd() # fileurlcsv <- "http://spark-public.s3.amazonaws.com/pdstatistics/data_sets/nesarc_pds.csv" # download.file(fileurlcsv, destfile="./nesarc_pds.csv") # list.files(".") # dateDownloaded <-date() # dateDownloaded nesarc <- read.csv("./nesarc_pds.csv") str(nesarc) # summary(nesarc) # sapply(nesarc[1, ], class) # sum(is.na(nesarc)) # table(is.na(nesarc)) library(Hmisc) library(RColorBrewer) library(reshape2) library(stringr) library(plyr) # # -------------------------------------------------------------------------------------------------------- # # # DEFINING THE VARIABLES UNDER STUDY # # # # study variables: Sex, average daily quantity of alcohol consumed and cigarettes smoked in past 12 months, and use of sedatives, tranquilizers, cannabis, opioids, amphetamines, cocaine, heroine, hallucinogens, and inhalants # codes: Sex, s2aq8b, s3aq3c1, S3bq1a1, S3bq1a2, S3bq1a3, S3bq1a4, S3bq1a5, S3bq1a6, S3bq1a7, S3bq1a8, S3bq1a9a #number of alcohol consumed in past 12 months table(is.na(nesarc$S2AQ8B)) table(nesarc$S2AQ8B) sum(table(nesarc$S2AQ8B)) #usual quantity of cigarettes smoked table(is.na(nesarc$S3AQ3C1)) table(nesarc$S3AQ3C1) sum(table(nesarc$S3AQ3C1)) # # -------------------------------------------------------------------------------------------------------- # without DATA MUNGING, it becomes weird, since binomials are coded differently [as 1 or 2], and NAs were assigned a numeric value [9] # here we are also only using LINEAR REGRESSION modeling, not accounting for the binomial distribution of some of these variables lmNoAdjust <- lm(MAJORDEPLIFE ~ SEX+ S2AQ8B+ S3AQ3C1+ S3BQ1A1+ S3BQ1A2+ S3BQ1A3+ S3BQ1A4+ S3BQ1A5+ S3BQ1A6+ S3BQ1A7+ S3BQ1A8+ S3BQ1A9A, data=nesarc) summary(lmNoAdjust) # Coefficients: # Estimate Std. Error t value Pr(>|t|) # (Intercept) 0.0347409 0.0223817 1.552 0.120639 # SEX 0.1533657 0.0072438 21.172 < 2e-16 *** # S2AQ8B 0.0001574 0.0003837 0.410 0.681605 # S3AQ3C1 0.0010271 0.0002495 4.117 3.86e-05 *** # S3BQ1A1 -0.0788230 0.0155627 -5.065 4.14e-07 *** # S3BQ1A2 -0.0113460 0.0163401 -0.694 0.487463 # S3BQ1A3 -0.0418769 0.0148434 -2.821 0.004791 ** # S3BQ1A4 -0.0030821 0.0137318 -0.224 0.822409 # S3BQ1A5 -0.0815162 0.0080903 -10.076 < 2e-16 *** # S3BQ1A6 -0.0115252 0.0133963 -0.860 0.389627 # S3BQ1A7 -0.0105396 0.0134567 -0.783 0.433512 # S3BQ1A8 0.0517945 0.0150253 3.447 0.000568 *** # S3BQ1A9A 0.1440555 0.0179928 8.006 1.28e-15 *** # # -------------------------------------------------------------------------------------------------------- # # # CREATE NEW DATAFRAME WITH FEWER VARIABLES # # # mddold <- data.frame(nesarc$MAJORDEPLIFE, nesarc$SEX, nesarc$S3AQ3B1, nesarc$S3AQ3C1, nesarc$S2AQ8A, nesarc$S2AQ8B, nesarc$S3BQ1A1, nesarc$S3BQ1A2, nesarc$S3BQ1A3, nesarc$S3BQ1A4, nesarc$S3BQ1A5, nesarc$S3BQ1A6, nesarc$S3BQ1A7, nesarc$S3BQ1A8, nesarc$S3BQ1A9A) str(mddold) names(mddold) <- c("mdd", "sex", "smokefreq", "smoke", "drinkfreq", "alcohol", "sedatives", "tranquilizers", "opioids", "amphetamines", "cannabis", "cocaine", "hallucinogens", "inhalants", "heroine") str(mddold) # # -------------------------------------------------------------------------------------------------------- # # # DATA MUNGING # # # # for the following variables: "sex", "sedatives", "tranquilizers", "cannabis", "opioids", "amphetamines", "cocaine", "heroine", "hallucinogens", "inhalants" # change the binomial setting [1,2] --> [1,0] # change [9] to [NA] # # # DATA MUNGING # # # # for the following variables: "alcohol", "smoke", "smokefreq", "drinkfreq" # change [99] for smoke quantity, drinks consumed to [NA] # change [99] for drinkfreq to [NA] # change [9] for smokefreq to [NA] mddold$sex[mddold$sex==2]=0 # sex=0 female; sex=1 male mddold$sedatives[mddold$sedatives==2]=0 mddold$tranquilizers[mddold$tranquilizers==2]=0 mddold$cannabis[mddold$cannabis==2]=0 mddold$opioids[mddold$opioids==2]=0 mddold$amphetamines[mddold$amphetamines==2]=0 mddold$cocaine[mddold$cocaine==2]=0 mddold$heroine[mddold$heroine==2]=0 mddold$hallucinogens[mddold$hallucinogens==2]=0 mddold$inhalants[mddold$inhalants==2]=0 mddold$sedatives[mddold$sedatives==9]=NA mddold$tranquilizers[mddold$tranquilizers==9]=NA mddold$cannabis[mddold$cannabis==9]=NA mddold$opioids[mddold$opioids==9]=NA mddold$amphetamines[mddold$amphetamines==9]=NA mddold$cocaine[mddold$cocaine==9]=NA mddold$heroine[mddold$heroine==9]=NA mddold$hallucinogens[mddold$hallucinogens==9]=NA mddold$inhalants[mddold$inhalants==9]=NA mddold$alcohol[mddold$alcohol==99]=NA mddold$smoke[mddold$smoke==99]=NA mddold$smokefreq[mddold$smokefreq==9]=NA mddold$drinkfreq[mddold$drinkfreq==99]=NA # # # ADD NEW VARIABLES: PACKYEARS and DRINKYEARS # # # mddold$smokefreqyr[mddold$smokefreq==1]= 364 mddold$smokefreqyr[mddold$smokefreq==2]= 286 mddold$smokefreqyr[mddold$smokefreq==3]= 182 mddold$smokefreqyr[mddold$smokefreq==4]= 78 mddold$smokefreqyr[mddold$smokefreq==5]= 30 mddold$smokefreqyr[mddold$smokefreq==6]= 1 mddold$smokefreqyr[mddold$smokefreq==NA]= NA mddold$alcoholfreqyr[mddold$drinkfreq==1]=364 mddold$alcoholfreqyr[mddold$drinkfreq==2]=286 mddold$alcoholfreqyr[mddold$drinkfreq==3]=182 mddold$alcoholfreqyr[mddold$drinkfreq==4]=104 mddold$alcoholfreqyr[mddold$drinkfreq==5]=52 mddold$alcoholfreqyr[mddold$drinkfreq==6]=30 mddold$alcoholfreqyr[mddold$drinkfreq==7]=12 mddold$alcoholfreqyr[mddold$drinkfreq==8]=9 mddold$alcoholfreqyr[mddold$drinkfreq==9]=4.5 mddold$alcoholfreqyr[mddold$drinkfreq==10]=1.5 mddold$alcoholfreqyr[mddold$drinkfreq==NA]=NA mddold$cigsperyear <- (mddold$smokefreqyr * mddold$smoke) mddold$swigsperyear <- (mddold$alcoholfreqyr * mddold$alcohol) summary(mddold$smokefreqyr) summary(mddold$alcoholfreqyr) summary(mddold$cigsperyear) summary(mddold$swigsperyear) table(mddold$smokefreqyr) table(mddold$alcoholfreqyr) table(mddold$cigsperyear) table(mddold$swigsperyear) mddnew <- mddold str(mddnew) # # -------------------------------------------------------------------------------------------------------- # FREQUENCY TABLES # ("mdd", "sex", "drinkyears", "packyears", "sedatives", "tranquilizers", "cannabis", "opioids", "amphetamines", "cocaine", "heroine", "hallucinogens", "inhalants") library(gmodels) CrossTable(mddnew$mdd, mddnew$sex) CrossTable(mddnew$mdd, mddnew$sedatives) CrossTable(mddnew$mdd, mddnew$tranquilizers) CrossTable(mddnew$mdd, mddnew$cannabis) CrossTable(mddnew$mdd, mddnew$opioids) CrossTable(mddnew$mdd, mddnew$amphetamines) CrossTable(mddnew$mdd, mddnew$cocaine) CrossTable(mddnew$mdd, mddnew$heroine) CrossTable(mddnew$mdd, mddnew$hallucinogens) CrossTable(mddnew$mdd, mddnew$inhalants) library(Hmisc) mddnew$g4cigs <- cut2(mddnew$cigsperyear, g = 4) CrossTable(mddnew$mdd, mddnew$g4cigs) # intervals # [ 1, 2002) | [2002, 4550) | [4550, 7644) | [7644,35672] mddnew$g4swigs <- cut2(mddnew$swigsperyear, g = 4) CrossTable(mddnew$mdd, mddnew$g4swigs) # intervals # [ 1.5, 10.5) | [ 10.5, 63.0) | [ 63.0, 360.0) | [360.0,35672.0] # # -------------------------------------------------------------------------------------------------------- # # # MULTIVARIATE GRAPHS FOR EXPLORATORY ANALYSIS # # # library(RColorBrewer) mypar <- function(a = 1, b = 1, brewer.n = 4, brewer.name = "RdYlGn", ...) { par(mar = c(2.5, 2.5, 1.6, 1.1), mgp = c(1.5, 0.5, 0)) par(mfrow = c(a, b), ...) palette(brewer.pal(brewer.n, brewer.name)) } cols4v1 <- brewer.pal(5, "YlGnBu") pal4v1 <- colorRampPalette(cols4v1) # <- dark blue cyan white gradient cols4v2 <- brewer.pal(5, "YlOrBr") pal4v2 <- colorRampPalette(cols4v2) # <- brown orange white gradient cols4v3 <- brewer.pal(5, "PuRd") pal4v3 <- colorRampPalette(cols4v3) # <- purple pink white gradient cols8 <- brewer.pal(8, "Set3") pal8 <- colorRampPalette(cols8) cols12 <- brewer.pal(12, "Set3") pal12 <- colorRampPalette(cols12) # create table for mdd vs cigs category mddvg4cigs = table(mddnew$mdd,mddnew$g4cigs) # To get the graph we want, we need to exchange the rows in this table mddvg4cigs = rbind(mddvg4cigs[2,],mddvg4cigs[1,]) # and turn them into percents (dividing by the num. of observations # in each cigs category) mddvg4cigs[1,]=mddvg4cigs[1,]/table(mddnew$g4cigs) mddvg4cigs[2,]=mddvg4cigs[2,]/table(mddnew$g4cigs) str(mddvg4cigs) # create table for mdd vs alcohol category mddvg4swigs = table(mddnew$mdd,mddnew$g4swigs) # To get the graph we want, we need to exchange the rows in this table mddvg4swigs = rbind(mddvg4swigs[2,],mddvg4swigs[1,]) # and turn them into percents (dividing by the num. of observations # in each swigs category) mddvg4swigs[1,]=mddvg4swigs[1,]/table(mddnew$g4swigs) mddvg4swigs[2,]=mddvg4swigs[2,]/table(mddnew$g4swigs) str(mddvg4swigs) # create table for mdd vs sex mddvsex = table(mddnew$mdd,mddnew$sex) # To get the graph we want, we need to exchange the rows in this table mddvsex = rbind(mddvsex[2,],mddvsex[1,]) # and turn them into percents (dividing by the num. of observations # in either con or sans sex) mddvsex[1,]=mddvsex[1,]/table(mddnew$sex) mddvsex[2,]=mddvsex[2,]/table(mddnew$sex) str(mddvsex) # create table for mdd vs sedatives mddvsedatives = table(mddnew$mdd,mddnew$sedatives) # To get the graph we want, we need to exchange the rows in this table mddvsedatives = rbind(mddvsedatives[2,],mddvsedatives[1,]) # and turn them into percents (dividing by the num. of observations # in either con or sans sedatives) mddvsedatives[1,]=mddvsedatives[1,]/table(mddnew$sedatives) mddvsedatives[2,]=mddvsedatives[2,]/table(mddnew$sedatives) str(mddvsedatives) # create table for mdd vs cannabis mddvcannabis = table(mddnew$mdd,mddnew$cannabis) # To get the graph we want, we need to exchange the rows in this table mddvcannabis = rbind(mddvcannabis[2,],mddvcannabis[1,]) # and turn them into percents (dividing by the num. of observations # in either con or sans cannabis) mddvcannabis[1,]=mddvcannabis[1,]/table(mddnew$cannabis) mddvcannabis[2,]=mddvcannabis[2,]/table(mddnew$cannabis) str(mddvcannabis) # create table for mdd vs opioids mddvopioids = table(mddnew$mdd,mddnew$opioids) # To get the graph we want, we need to exchange the rows in this table mddvopioids = rbind(mddvopioids[2,],mddvopioids[1,]) # and turn them into percents (dividing by the num. of observations # in either con or sans opioids) mddvopioids[1,]=mddvopioids[1,]/table(mddnew$opioids) mddvopioids[2,]=mddvopioids[2,]/table(mddnew$opioids) str(mddvopioids) # create table for mdd vs tranquilizers mddvtranquilizers = table(mddnew$mdd,mddnew$tranquilizers) # To get the graph we want, we need to exchange the rows in this table mddvtranquilizers = rbind(mddvtranquilizers[2,],mddvtranquilizers[1,]) # and turn them into percents (dividing by the num. of observations # in either con or sans tranquilizers) mddvtranquilizers[1,]=mddvtranquilizers[1,]/table(mddnew$tranquilizers) mddvtranquilizers[2,]=mddvtranquilizers[2,]/table(mddnew$tranquilizers) str(mddvtranquilizers) # create table for mdd vs amphetamines mddvamphetamines = table(mddnew$mdd,mddnew$amphetamines) # To get the graph we want, we need to exchange the rows in this table mddvamphetamines = rbind(mddvamphetamines[2,],mddvamphetamines[1,]) # and turn them into percents (dividing by the num. of observations # in either con or sans amphetamines) mddvamphetamines[1,]=mddvamphetamines[1,]/table(mddnew$amphetamines) mddvamphetamines[2,]=mddvamphetamines[2,]/table(mddnew$amphetamines) str(mddvamphetamines) # create table for mdd vs cocaine mddvcocaine = table(mddnew$mdd,mddnew$cocaine) # To get the graph we want, we need to exchange the rows in this table mddvcocaine = rbind(mddvcocaine[2,],mddvcocaine[1,]) # and turn them into percents (dividing by the num. of observations # in either con or sans cocaine) mddvcocaine[1,]=mddvcocaine[1,]/table(mddnew$cocaine) mddvcocaine[2,]=mddvcocaine[2,]/table(mddnew$cocaine) str(mddvcocaine) # create table for mdd vs heroine mddvheroine = table(mddnew$mdd,mddnew$heroine) # To get the graph we want, we need to exchange the rows in this table mddvheroine = rbind(mddvheroine[2,],mddvheroine[1,]) # and turn them into percents (dividing by the num. of observations # in either con or sans heroine) mddvheroine[1,]=mddvheroine[1,]/table(mddnew$heroine) mddvheroine[2,]=mddvheroine[2,]/table(mddnew$heroine) str(mddvheroine) # create table for mdd vs hallucinogens mddvhallucinogens = table(mddnew$mdd,mddnew$hallucinogens) # To get the graph we want, we need to exchange the rows in this table mddvhallucinogens = rbind(mddvhallucinogens[2,],mddvhallucinogens[1,]) # and turn them into percents (dividing by the num. of observations # in either con or sans hallucinogens) mddvhallucinogens[1,]=mddvhallucinogens[1,]/table(mddnew$hallucinogens) mddvhallucinogens[2,]=mddvhallucinogens[2,]/table(mddnew$hallucinogens) str(mddvhallucinogens) # create table for mdd vs inhalants mddvinhalants = table(mddnew$mdd,mddnew$inhalants) # To get the graph we want, we need to exchange the rows in this table mddvinhalants = rbind(mddvinhalants[2,],mddvinhalants[1,]) # and turn them into percents (dividing by the num. of observations # in either con or sans inhalants) mddvinhalants[1,]=mddvinhalants[1,]/table(mddnew$inhalants) mddvinhalants[2,]=mddvinhalants[2,]/table(mddnew$inhalants) str(mddvinhalants) mypar(mfrow = c(2,2)) # MDD diagnosis {RESPONSE} by Estimated Cigarette Use per Year {EXPLANATORY} among all Adults in the NESARC Study bp_mddvg4cigs <- barplot(mddvg4cigs[1,], col=unique(mddnew$g4cigs), xlab="cigarettes per year", ylab="diagnosed depression", cex.axis=0.8) # MDD diagnosis {RESPONSE} by Estimated Cigarette Use per Year {EXPLANATORY} among all Adults in the NESARC Study bp_mddvg4cigs <- barplot(mddvg4cigs, col=unique(mddnew$g4cigs), xlab="cigarettes per year", ylab="diagnosed depression", cex.axis=0.8) # MDD diagnosis {RESPONSE} by Estimated Alcohol Consumed per Year {EXPLANATORY} among all Adults in the NESARC Study bp_mddvg4swigs <- barplot(mddvg4swigs[1,], col=unique(mddnew$g4swigs), xlab="alcohol per year", ylab="diagnosed depression", cex.axis=0.8) # MDD diagnosis {RESPONSE} by Estimated Alcohol Consumed per Year {EXPLANATORY} among all Adults in the NESARC Study bp_mddvg4swigs <- barplot(mddvg4swigs, col=unique(mddnew$g4swigs), xlab="alcohol per year", ylab="diagnosed depression", cex.axis=0.8) #dev.copy2pdf(file="mdd_cigarettes_alcohol.pdf", height =8, width = 11) mypar(mfrow = c(2, 5)) # MDD diagnosis {RESPONSE} by Biological Sex {EXPLANATORY} among all Adults in the NESARC Study bp_mddvsex <- barplot(mddvsex, col=pal4v1(4), xlab="biological sex, 0 - female, 1 - male", ylab="diagnosed depression") # MDD diagnosis {RESPONSE} by Sedatives Use {EXPLANATORY} among all Adults in the NESARC Study bp_mddvsedatives <- barplot(mddvsedatives, col=pal4v2(4), xlab="sedative use", ylab="diagnosed depression") # MDD diagnosis {RESPONSE} by Cannabis Use {EXPLANATORY} among all Adults in the NESARC Study bp_mddvcannabis <- barplot(mddvcannabis, col=pal4v3(4), xlab="cannabis use", ylab="diagnosed depression") # MDD diagnosis {RESPONSE} by tranquilizers use {EXPLANATORY} among all Adults in the NESARC Study bp_mddvtranquilizers <- barplot(mddvtranquilizers, col=pal8(4), xlab="tranquilizers use", ylab="diagnosed depression") # MDD diagnosis {RESPONSE} by opioids use {EXPLANATORY} among all Adults in the NESARC Study bp_mddvopioids <- barplot(mddvopioids, col=pal12(4), xlab="opioids use", ylab="diagnosed depression") # MDD diagnosis {RESPONSE} by cocaine use {EXPLANATORY} among all Adults in the NESARC Study bp_mddvcocaine <- barplot(mddvcocaine, col=pal12(4), xlab="cocaine use", ylab="diagnosed depression") # MDD diagnosis {RESPONSE} by amphetamines use {EXPLANATORY} among all Adults in the NESARC Study bp_mddvamphetamines <- barplot(mddvamphetamines, col=pal8(4), xlab="amphetamines use", ylab="diagnosed depression") # MDD diagnosis {RESPONSE} by heroine use {EXPLANATORY} among all Adults in the NESARC Study bp_mddvheroine <- barplot(mddvheroine, col=pal4v3(4), xlab="heroine use", ylab="diagnosed depression") # MDD diagnosis {RESPONSE} by hallucinogens use {EXPLANATORY} among all Adults in the NESARC Study bp_mddvhallucinogens <- barplot(mddvhallucinogens, col=pal4v2(4), xlab="hallucinogens use", ylab="diagnosed depression") # MDD diagnosis {RESPONSE} by inhalants use {EXPLANATORY} among all Adults in the NESARC Study bp_mddvinhalants <- barplot(mddvinhalants, col=pal4v1(4), xlab="inhalants use", ylab="diagnosed depression") #dev.copy2pdf(file="mdd_sex_substances.pdf", height =8, width = 11) mypar(mfrow=c(1,2)) boxplot(mddnew$cigsperyear, col="#FE9929", xlab = "# of cigarettes smoked per year", cex=1.5) boxplot(mddnew$swigsperyear, log="y", col="#41B6C4", xlab = "# of alcohol consumed per year", cex=1.5) #dev.copy2pdf(file="boxplots cigs and swigs.pdf", height =4, width = 7) # # -------------------------------------------------------------------------------------------------------- # # MODELING # # # SEPARATE MODELS FOR FEMALES AND MALES males <- mddnew[[2]] == 1 females <- mddnew[[2]] == 0 mddnewmales <- mddnew[males,] mddnewfemales <- mddnew[females,] str(mddnewmales) str(mddnewfemales) logsubstancesall <- glm(mdd ~ as.factor(g4swigs) + as.factor(g4cigs) + sedatives + tranquilizers + cannabis + opioids + amphetamines + cocaine + heroine + hallucinogens + inhalants, data=mddnew, family="binomial") logsubstancesmales <- glm(mdd ~ as.factor(g4swigs) + as.factor(g4cigs) + sedatives + tranquilizers + cannabis + opioids + amphetamines + cocaine + heroine + hallucinogens + inhalants, data=mddnewmales, family="binomial") logsubstancesfemales <- glm(mdd ~ as.factor(g4swigs) + as.factor(g4cigs) + sedatives + tranquilizers + cannabis + opioids + amphetamines + cocaine + heroine + hallucinogens + inhalants, data=mddnewfemales, family="binomial") summary(logsubstancesall) exp(logsubstancesall$coeff) exp(confint(logsubstancesall)) plot (mddnew$mdd, logsubstancesall$fitted) anova(logsubstancesall,test="Chisq") summary(logsubstancesmales) exp(logsubstancesmales$coeff) exp(confint(logsubstancesmales)) plot (mddnewmales$mdd, logsubstancesmales$fitted) anova(logsubstancesmales,test="Chisq") # Call: # glm(formula = mdd ~ as.factor(g4swigs) + as.factor(g4cigs) + # sedatives + tranquilizers + cannabis + opioids + amphetamines + # cocaine + heroine + hallucinogens + inhalants, family = "binomial", # data = mddnewmales) # # Deviance Residuals: # Min 1Q Median 3Q Max # -1.3316 -0.5962 -0.4860 -0.4472 2.1951 # # Coefficients: # Estimate Std. Error z value Pr(>|z|) # (Intercept) -2.1097030 0.1136347 -18.566 < 2e-16 *** # as.factor(g4swigs)[ 10.5, 63.0) 0.1165468 0.1174018 0.993 0.320848 # as.factor(g4swigs)[ 63.0, 360.0) -0.1015568 0.1145740 -0.886 0.375409 # as.factor(g4swigs)[360.0,35672.0] -0.1425212 0.1073813 -1.327 0.184428 # as.factor(g4cigs)[2002, 4550) -0.0627617 0.1074292 -0.584 0.559076 # as.factor(g4cigs)[4550, 7644) 0.0330148 0.0902443 0.366 0.714486 # as.factor(g4cigs)[7644,35672] 0.3729660 0.1058383 3.524 0.000425 *** <-- this is slightly significant # sedatives 0.5647761 0.1393265 4.054 5.04e-05 *** <-- this is slightly significant # tranquilizers 0.0351090 0.1557427 0.225 0.821645 # cannabis 0.5819275 0.0845202 6.885 5.78e-12 *** <-- this is slightly significant # opioids 0.3639522 0.1264450 2.878 0.003998 ** <-- this is slightly significant # amphetamines 0.1981210 0.1285637 1.541 0.123308 # cocaine 0.0643056 0.1195646 0.538 0.590693 # heroine -0.0002183 0.2815689 -0.001 0.999382 # hallucinogens 0.1674107 0.1238734 1.351 0.176546 # inhalants 0.1386452 0.1516635 0.914 0.360631 # --- # Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 # (Dispersion parameter for binomial family taken to be 1) # # Null deviance: 5767.9 on 6544 degrees of freedom # Residual deviance: 5442.7 on 6529 degrees of freedom # (11973 observations deleted due to missingness) # AIC: 5474.7 # Number of Fisher Scoring iterations: 4 # > exp(logsubstancesmales$coeff) # (Intercept) as.factor(g4swigs)[ 10.5, 63.0) # 0.1212740 1.1236101 # as.factor(g4swigs)[ 63.0, 360.0) as.factor(g4swigs)[360.0,35672.0] # 0.9034299 0.8671692 # as.factor(g4cigs)[2002, 4550) as.factor(g4cigs)[4550, 7644) # 0.9391673 1.0335659 # as.factor(g4cigs)[7644,35672] sedatives # 1.4520350 1.7590538 # tranquilizers cannabis # 1.0357326 1.7894844 # opioids amphetamines # 1.4390054 1.2191099 # cocaine heroine # 1.0664182 0.9997818 # hallucinogens inhalants # 1.1822398 1.1487165 # > exp(confint(logsubstancesmales)) # Waiting for profiling to be done... # 2.5 % 97.5 % # (Intercept) 0.09672887 0.1510365 # as.factor(g4swigs)[ 10.5, 63.0) 0.89350182 1.4160715 # as.factor(g4swigs)[ 63.0, 360.0) 0.72254676 1.1325033 # as.factor(g4swigs)[360.0,35672.0] 0.70381227 1.0724474 # as.factor(g4cigs)[2002, 4550) 0.76016557 1.1585051 # as.factor(g4cigs)[4550, 7644) 0.86652156 1.2344479 # as.factor(g4cigs)[7644,35672] 1.17957468 1.7864351 <-- this is slightly significant # sedatives 1.33680824 2.3088374 <-- this is slightly significant # tranquilizers 0.76170329 1.4030169 # cannabis 1.51531707 2.1107442 <-- this is slightly significant # opioids 1.12075291 1.8402787 <-- this is slightly significant # amphetamines 0.94609989 1.5664145 # cocaine 0.84226946 1.3461298 # heroine 0.57008024 1.7265223 # hallucinogens 0.92610582 1.5053609 # inhalants 0.85095214 1.5426781 # > plot (mddnewmales$mdd, logsubstancesmales$fitted) # Error in xy.coords(x, y, xlabel, ylabel, log) : # 'x' and 'y' lengths differ # > anova(logsubstancesmales,test="Chisq") # Analysis of Deviance Table # Model: binomial, link: logit # Response: mdd # Terms added sequentially (first to last) # Df Deviance Resid. Df Resid. Dev Pr(>Chi) # NULL 6544 5767.9 # as.factor(g4swigs) 3 4.398 6541 5763.5 0.2215738 # as.factor(g4cigs) 3 19.022 6538 5744.4 0.0002705 *** <-- this is significant # sedatives 1 175.136 6537 5569.3 < 2.2e-16 *** <-- this is significant # tranquilizers 1 14.750 6536 5554.6 0.0001227 *** <-- this is significant # cannabis 1 89.467 6535 5465.1 < 2.2e-16 *** <-- this is significant # opioids 1 12.316 6534 5452.8 0.0004492 *** <-- this is significant # amphetamines 1 5.754 6533 5447.0 0.0164489 * <-- this is significant # cocaine 1 1.310 6532 5445.7 0.2523128 # heroine 1 0.025 6531 5445.7 0.8751546 # hallucinogens 1 2.193 6530 5443.5 0.1386587 # inhalants 1 0.829 6529 5442.7 0.3626344 # --- # Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 summary(logsubstancesfemales) exp(logsubstancesfemales$coeff) exp(confint(logsubstancesfemales)) plot (mddnewfemales$mdd, logsubstancesfemales$fitted) anova(logsubstancesfemales,test="Chisq") # > summary(logsubstancesfemales) # # Call: # glm(formula = mdd ~ as.factor(g4swigs) + as.factor(g4cigs) + # sedatives + tranquilizers + cannabis + opioids + amphetamines + # cocaine + heroine + hallucinogens + inhalants, family = "binomial", # data = mddnewfemales) # # Deviance Residuals: # Min 1Q Median 3Q Max # -1.7306 -0.8047 -0.7157 1.2613 1.7912 # # Coefficients: # Estimate Std. Error z value Pr(>|z|) # (Intercept) -1.23135 0.07360 -16.731 < 2e-16 *** # as.factor(g4swigs)[ 10.5, 63.0) -0.10289 0.07709 -1.335 0.181991 # as.factor(g4swigs)[ 63.0, 360.0) -0.07070 0.07920 -0.893 0.372032 # as.factor(g4swigs)[360.0,35672.0] -0.11143 0.08503 -1.310 0.190061 # as.factor(g4cigs)[2002, 4550) 0.08761 0.07727 1.134 0.256842 # as.factor(g4cigs)[4550, 7644) 0.26987 0.07261 3.716 0.000202 *** # as.factor(g4cigs)[7644,35672] 0.51903 0.10521 4.934 8.08e-07 *** # sedatives 0.23058 0.13434 1.716 0.086080 . # tranquilizers 0.25377 0.14481 1.752 0.079695 . # cannabis 0.60725 0.06880 8.827 < 2e-16 *** # opioids 0.44915 0.12660 3.548 0.000388 *** # amphetamines 0.34957 0.12553 2.785 0.005357 ** # cocaine 0.10326 0.11098 0.931 0.352105 # heroine 0.30044 0.46536 0.646 0.518537 # hallucinogens -0.03703 0.11994 -0.309 0.757524 # inhalants -0.22342 0.19757 -1.131 0.258134 # --- # Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 # (Dispersion parameter for binomial family taken to be 1) # # Null deviance: 7461.1 on 6040 degrees of freedom # Residual deviance: 7153.5 on 6025 degrees of freedom # (18534 observations deleted due to missingness) # AIC: 7185.5 # # Number of Fisher Scoring iterations: 4 # > exp(logsubstancesfemales$coeff) # (Intercept) as.factor(g4swigs)[ 10.5, 63.0) as.factor(g4swigs)[ 63.0, 360.0) # 0.2918978 0.9022268 0.9317417 # as.factor(g4swigs)[360.0,35672.0] as.factor(g4cigs)[2002, 4550) as.factor(g4cigs)[4550, 7644) # 0.8945575 1.0915670 1.3097916 # as.factor(g4cigs)[7644,35672] sedatives tranquilizers # 1.6803973 1.2593351 1.2888798 # cannabis opioids amphetamines # 1.8353745 1.5669720 1.4184636 # cocaine heroine hallucinogens # 1.1087842 1.3504529 0.9636485 # inhalants # 0.7997824 # > exp(confint(logsubstancesfemales)) # Waiting for profiling to be done... # 2.5 % 97.5 % # (Intercept) 0.2524219 0.3368565 # as.factor(g4swigs)[ 10.5, 63.0) 0.7755918 1.0492891 # as.factor(g4swigs)[ 63.0, 360.0) 0.7976125 1.0880415 # as.factor(g4swigs)[360.0,35672.0] 0.7568787 1.0563681 # as.factor(g4cigs)[2002, 4550) 0.9381185 1.2700609 # as.factor(g4cigs)[4550, 7644) 1.1362570 1.5104773 # as.factor(g4cigs)[7644,35672] 1.3663559 2.0641066 # sedatives 0.9667310 1.6374606 # tranquilizers 0.9694174 1.7109332 # cannabis 1.6035308 2.0999779 # opioids 1.2219634 2.0078102 # amphetamines 1.1087079 1.8140419 # cocaine 0.8913747 1.3774276 # heroine 0.5462273 3.4608405 # hallucinogens 0.7609479 1.2179831 # inhalants 0.5414240 1.1761289 # > anova(logsubstancesfemales,test="Chisq") # Analysis of Deviance Table # Model: binomial, link: logit # Response: mdd # Terms added sequentially (first to last) # Df Deviance Resid. Df Resid. Dev Pr(>Chi) # NULL 6040 7461.1 # as.factor(g4swigs) 3 1.950 6037 7459.2 0.582951 # as.factor(g4cigs) 3 34.849 6034 7424.3 1.311e-07 *** # sedatives 1 86.558 6033 7337.8 < 2.2e-16 *** # tranquilizers 1 30.171 6032 7307.6 3.955e-08 *** # cannabis 1 127.106 6031 7180.5 < 2.2e-16 *** # opioids 1 15.441 6030 7165.1 8.511e-05 *** # amphetamines 1 8.984 6029 7156.1 0.002723 ** # cocaine 1 0.711 6028 7155.4 0.399268 # heroine 1 0.288 6027 7155.1 0.591425 # hallucinogens 1 0.252 6026 7154.8 0.615423 # inhalants 1 1.286 6025 7153.5 0.256804 # --- # Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1