Project
Julianna Fuentes
5/11/2021
fundraising = readRDS("fundraising.rds")
summary(fundraising)## zipconvert2 zipconvert3 zipconvert4 zipconvert5 homeowner num_child
## No :2352 Yes: 551 No :2357 No :1846 Yes:2312 Min. :1.000
## Yes: 648 No :2449 Yes: 643 Yes:1154 No : 688 1st Qu.:1.000
## Median :1.000
## Mean :1.069
## 3rd Qu.:1.000
## Max. :5.000
## income female wealth home_value med_fam_inc
## Min. :1.000 Yes:1831 Min. :0.000 Min. : 0.0 Min. : 0.0
## 1st Qu.:3.000 No :1169 1st Qu.:5.000 1st Qu.: 554.8 1st Qu.: 278.0
## Median :4.000 Median :8.000 Median : 816.5 Median : 355.0
## Mean :3.899 Mean :6.396 Mean :1143.3 Mean : 388.4
## 3rd Qu.:5.000 3rd Qu.:8.000 3rd Qu.:1341.2 3rd Qu.: 465.0
## Max. :7.000 Max. :9.000 Max. :5945.0 Max. :1500.0
## avg_fam_inc pct_lt15k num_prom lifetime_gifts
## Min. : 0.0 Min. : 0.00 Min. : 11.00 Min. : 15.0
## 1st Qu.: 318.0 1st Qu.: 5.00 1st Qu.: 29.00 1st Qu.: 45.0
## Median : 396.0 Median :12.00 Median : 48.00 Median : 81.0
## Mean : 432.3 Mean :14.71 Mean : 49.14 Mean : 110.7
## 3rd Qu.: 516.0 3rd Qu.:21.00 3rd Qu.: 65.00 3rd Qu.: 135.0
## Max. :1331.0 Max. :90.00 Max. :157.00 Max. :5674.9
## largest_gift last_gift months_since_donate time_lag
## Min. : 5.00 Min. : 0.00 Min. :17.00 Min. : 0.000
## 1st Qu.: 10.00 1st Qu.: 7.00 1st Qu.:29.00 1st Qu.: 3.000
## Median : 15.00 Median : 10.00 Median :31.00 Median : 5.000
## Mean : 16.65 Mean : 13.48 Mean :31.13 Mean : 6.876
## 3rd Qu.: 20.00 3rd Qu.: 16.00 3rd Qu.:34.00 3rd Qu.: 9.000
## Max. :1000.00 Max. :219.00 Max. :37.00 Max. :77.000
## avg_gift target
## Min. : 2.139 Donor :1499
## 1st Qu.: 6.333 No Donor:1501
## Median : 9.000
## Mean : 10.669
## 3rd Qu.: 12.800
## Max. :122.167The dataset contains 3,000 observations. The response variabe has two levels, Donor and No Donor. Each level has approximately 50% of the total observations, the result of weighted sampling.
Using pairs plot to examinethe predictors to see if there are any associations between the predictors and the response variable.
pairs(fundraising[5:17])
med_fam_inc, avg_fam_inc, and home_value have a positive correlation between one another and a negative correlation with pct_lt15k.
pairs(fundraising[16:21])
str(fundraising)## Classes 'tbl_df', 'tbl' and 'data.frame': 3000 obs. of 21 variables:
## $ zipconvert2 : Factor w/ 2 levels "No","Yes": 2 1 1 1 1 1 1 2 1 2 ...
## $ zipconvert3 : Factor w/ 2 levels "Yes","No": 2 2 2 1 1 2 2 2 2 2 ...
## $ zipconvert4 : Factor w/ 2 levels "No","Yes": 1 1 1 1 1 1 2 1 1 1 ...
## $ zipconvert5 : Factor w/ 2 levels "No","Yes": 1 2 2 1 1 2 1 1 2 1 ...
## $ homeowner : Factor w/ 2 levels "Yes","No": 1 2 1 1 1 1 1 1 1 1 ...
## $ num_child : num 1 2 1 1 1 1 1 1 1 1 ...
## $ income : num 1 5 3 4 4 4 4 4 4 1 ...
## $ female : Factor w/ 2 levels "Yes","No": 2 1 2 2 1 1 2 1 1 1 ...
## $ wealth : num 7 8 4 8 8 8 5 8 8 5 ...
## $ home_value : num 698 828 1471 547 482 ...
## $ med_fam_inc : num 422 358 484 386 242 450 333 458 541 203 ...
## $ avg_fam_inc : num 463 376 546 432 275 498 388 533 575 271 ...
## $ pct_lt15k : num 4 13 4 7 28 5 16 8 11 39 ...
## $ num_prom : num 46 32 94 20 38 47 51 21 66 73 ...
## $ lifetime_gifts : num 94 30 177 23 73 139 63 26 108 161 ...
## $ largest_gift : num 12 10 10 11 10 20 15 16 12 6 ...
## $ last_gift : num 12 5 8 11 10 20 10 16 7 3 ...
## $ months_since_donate: num 34 29 30 30 31 37 37 30 31 32 ...
## $ time_lag : num 6 7 3 6 3 3 8 6 1 7 ...
## $ avg_gift : num 9.4 4.29 7.08 7.67 7.3 ...
## $ target : Factor w/ 2 levels "Donor","No Donor": 1 1 2 2 1 1 1 2 1 1 ...numerical variables are - num_child income, wealth, home_value, med_fam_inc, avg_fam_inc, ptc_lt15k, num_prom, lifetime_gifts, largest_gift, months_since_donate, time_lag, and avg_gift. categorical variables are zipconvert2, zipconvert3, zipconvert4, zipconvert5, homeowner, female, and target.
count.donors = subset(fundraising, target == "Donor")
library(plyr)library(dplyr)##
## Attaching package: 'dplyr'## The following objects are masked from 'package:plyr':
##
## arrange, count, desc, failwith, id, mutate, rename, summarise,
## summarize## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unionfundNew = fundraising %>%
mutate(
female_new = case_when(
female == "Yes" ~ 1,
female == "No" ~ 0),
target_bin = case_when(
target == "Donor" ~ 1,
target == "No Donor" ~ 0),
home_bin = case_when(
homeowner == "Yes" ~ 1,
homeowner == "No" ~ 0)
)fundraiser = select(fundNew, -female, - target, - homeowner)
fundNum = fundraiser[, c(21, 5, 6, 7, 19, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 20)]
str(fundNum)## tibble [3,000 × 17] (S3: tbl_df/tbl/data.frame)
## $ home_bin : num [1:3000] 1 0 1 1 1 1 1 1 1 1 ...
## $ num_child : num [1:3000] 1 2 1 1 1 1 1 1 1 1 ...
## $ income : num [1:3000] 1 5 3 4 4 4 4 4 4 1 ...
## $ wealth : num [1:3000] 7 8 4 8 8 8 5 8 8 5 ...
## $ female_new : num [1:3000] 0 1 0 0 1 1 0 1 1 1 ...
## $ home_value : num [1:3000] 698 828 1471 547 482 ...
## $ med_fam_inc : num [1:3000] 422 358 484 386 242 450 333 458 541 203 ...
## $ avg_fam_inc : num [1:3000] 463 376 546 432 275 498 388 533 575 271 ...
## $ pct_lt15k : num [1:3000] 4 13 4 7 28 5 16 8 11 39 ...
## $ num_prom : num [1:3000] 46 32 94 20 38 47 51 21 66 73 ...
## $ lifetime_gifts : num [1:3000] 94 30 177 23 73 139 63 26 108 161 ...
## $ largest_gift : num [1:3000] 12 10 10 11 10 20 15 16 12 6 ...
## $ last_gift : num [1:3000] 12 5 8 11 10 20 10 16 7 3 ...
## $ months_since_donate: num [1:3000] 34 29 30 30 31 37 37 30 31 32 ...
## $ time_lag : num [1:3000] 6 7 3 6 3 3 8 6 1 7 ...
## $ avg_gift : num [1:3000] 9.4 4.29 7.08 7.67 7.3 ...
## $ target_bin : num [1:3000] 1 1 0 0 1 1 1 0 1 1 ...dataset fundNum now consists of 16 numerical predictors, and 1 numerical response variable
cor(fundNum)## home_bin num_child income wealth
## home_bin 1.0000000000 0.045246221 0.320470476 0.065770325
## num_child 0.0452462210 1.000000000 0.091893089 0.060175537
## income 0.3204704764 0.091893089 1.000000000 0.208993101
## wealth 0.0657703248 0.060175537 0.208993101 1.000000000
## female_new 0.0014784437 -0.029596832 -0.043813261 -0.029384095
## home_value 0.1187277614 -0.011964229 0.291973494 0.261161145
## med_fam_inc 0.1382760204 0.046961647 0.367505334 0.377763371
## avg_fam_inc 0.1334984472 0.047261395 0.378585352 0.385892299
## pct_lt15k -0.1375411145 -0.031717891 -0.283191234 -0.375145585
## num_prom 0.0032595805 -0.086432604 -0.069008634 -0.412117770
## lifetime_gifts -0.0281638939 -0.050954766 -0.019565470 -0.225473319
## largest_gift -0.0333386920 -0.017554416 0.033180760 -0.025276518
## last_gift -0.0008854337 -0.012948678 0.109592754 0.052591311
## months_since_donate 0.0213538579 -0.005563603 0.077238810 0.033713981
## time_lag 0.0262871410 -0.006069356 -0.001545727 -0.066421329
## avg_gift -0.0083209968 -0.019688680 0.124055750 0.091078754
## target_bin 0.0265945279 -0.042348253 0.035953287 0.003114465
## female_new home_value med_fam_inc avg_fam_inc
## home_bin 0.001478444 0.1187277614 0.138276020 0.133498447
## num_child -0.029596832 -0.0119642286 0.046961647 0.047261395
## income -0.043813261 0.2919734944 0.367505334 0.378585352
## wealth -0.029384095 0.2611611450 0.377763371 0.385892299
## female_new 1.000000000 -0.0209758272 -0.023450114 -0.025236799
## home_value -0.020975827 1.0000000000 0.738153074 0.752569002
## med_fam_inc -0.023450114 0.7381530742 1.000000000 0.972271285
## avg_fam_inc -0.025236799 0.7525690021 0.972271285 1.000000000
## pct_lt15k 0.055176372 -0.3990861577 -0.665362675 -0.680284797
## num_prom 0.038218180 -0.0645138583 -0.050782705 -0.057311385
## lifetime_gifts 0.037073694 -0.0240737013 -0.035245827 -0.040327155
## largest_gift 0.001381799 0.0564942757 0.047032066 0.043103937
## last_gift -0.046359831 0.1588576542 0.135976003 0.131378624
## months_since_donate -0.045056823 0.0234285142 0.032336691 0.031268594
## time_lag -0.008049159 0.0006789113 0.015202043 0.024340381
## avg_gift -0.074529893 0.1687736865 0.137162758 0.131758434
## target_bin 0.024757404 0.0215691141 0.008036116 0.003177139
## pct_lt15k num_prom lifetime_gifts largest_gift
## home_bin -0.1375411145 0.00325958 -0.02816389 -0.033338692
## num_child -0.0317178911 -0.08643260 -0.05095477 -0.017554416
## income -0.2831912335 -0.06900863 -0.01956547 0.033180760
## wealth -0.3751455847 -0.41211777 -0.22547332 -0.025276518
## female_new 0.0551763722 0.03821818 0.03707369 0.001381799
## home_value -0.3990861577 -0.06451386 -0.02407370 0.056494276
## med_fam_inc -0.6653626748 -0.05078270 -0.03524583 0.047032066
## avg_fam_inc -0.6802847967 -0.05731139 -0.04032716 0.043103937
## pct_lt15k 1.0000000000 0.03777518 0.05961881 -0.007882936
## num_prom 0.0377751828 1.00000000 0.53861957 0.113810342
## lifetime_gifts 0.0596188059 0.53861957 1.00000000 0.507262313
## largest_gift -0.0078829361 0.11381034 0.50726231 1.000000000
## last_gift -0.0617521213 -0.05586809 0.20205827 0.447236933
## months_since_donate -0.0090145584 -0.28232212 -0.14462186 0.019789633
## time_lag -0.0199114896 0.11962322 0.03854575 0.039977035
## avg_gift -0.0624808920 -0.14725094 0.18232435 0.474830096
## target_bin -0.0007592833 0.06836599 0.01962693 -0.017783355
## last_gift months_since_donate time_lag
## home_bin -0.0008854337 0.021353858 0.0262871410
## num_child -0.0129486780 -0.005563603 -0.0060693555
## income 0.1095927542 0.077238810 -0.0015457272
## wealth 0.0525913108 0.033713981 -0.0664213294
## female_new -0.0463598310 -0.045056823 -0.0080491595
## home_value 0.1588576542 0.023428514 0.0006789113
## med_fam_inc 0.1359760028 0.032336691 0.0152020426
## avg_fam_inc 0.1313786241 0.031268594 0.0243403812
## pct_lt15k -0.0617521213 -0.009014558 -0.0199114896
## num_prom -0.0558680871 -0.282322122 0.1196232155
## lifetime_gifts 0.2020582715 -0.144621862 0.0385457538
## largest_gift 0.4472369329 0.019789633 0.0399770354
## last_gift 1.0000000000 0.186715010 0.0751112090
## months_since_donate 0.1867150099 1.000000000 0.0155284995
## time_lag 0.0751112090 0.015528499 1.0000000000
## avg_gift 0.8663999778 0.189110799 0.0700816428
## target_bin -0.0777208200 -0.133813301 0.0097457015
## avg_gift target_bin
## home_bin -0.008320997 0.0265945279
## num_child -0.019688680 -0.0423482529
## income 0.124055750 0.0359532869
## wealth 0.091078754 0.0031144649
## female_new -0.074529893 0.0247574043
## home_value 0.168773687 0.0215691141
## med_fam_inc 0.137162758 0.0080361157
## avg_fam_inc 0.131758434 0.0031771394
## pct_lt15k -0.062480892 -0.0007592833
## num_prom -0.147250943 0.0683659889
## lifetime_gifts 0.182324349 0.0196269259
## largest_gift 0.474830096 -0.0177833547
## last_gift 0.866399978 -0.0777208200
## months_since_donate 0.189110799 -0.1338133012
## time_lag 0.070081643 0.0097457015
## avg_gift 1.000000000 -0.0756630051
## target_bin -0.075663005 1.0000000000avg_fam_inc, home_value, and med_fam_inc all have a positive correlation coefficient greater than 0.7, concluding thee is evidence of collinearity between them.
par(mfrow = c(1,2))
boxplot(fundNum$lifetime_gifts ~ fundNum$target_bin, data = fundNum)
boxplot(fundNum$largest_gift ~ fundNum$target_bin, data = fundNum)
par(mfrow = c(2,2))
boxplot(fundNum$last_gift ~ fundNum$target_bin, data = fundNum)
boxplot(fundNum$months_since_donate ~ fundNum$target_bin, data = fundNew)
boxplot(fundNum$time_lag ~ fundNum$target_bin, data = fundNum)
boxplot(fundNum$avg_gift ~ fundNum$target_bin, data = fundNum)
box plots show the distribution of the variables associated with the type of gifts recieved between Donor and Non-Donor.
library(ISLR)
set.seed(12345)
train = sample(3000, 2400)
train80 = fundNum[train,]
test20 = fundNum[-train,]
fund.test = fundNew[-train,]$targetfund.glm1 = glm(target_bin ~ home_bin + num_child + income + female_new + wealth + med_fam_inc + pct_lt15k + num_prom + lifetime_gifts + largest_gift + last_gift + months_since_donate + time_lag + avg_gift, data = train80, family = binomial)
summary(fund.glm1)##
## Call:
## glm(formula = target_bin ~ home_bin + num_child + income + female_new +
## wealth + med_fam_inc + pct_lt15k + num_prom + lifetime_gifts +
## largest_gift + last_gift + months_since_donate + time_lag +
## avg_gift, family = binomial, data = train80)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.6896 -1.1716 0.7697 1.1434 1.8349
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.483e+00 4.933e-01 3.007 0.00264 **
## home_bin 1.598e-01 1.054e-01 1.516 0.12965
## num_child -3.384e-01 1.275e-01 -2.655 0.00792 **
## income 5.197e-02 2.862e-02 1.816 0.06934 .
## female_new 2.470e-02 8.577e-02 0.288 0.77333
## wealth 1.998e-02 1.986e-02 1.006 0.31459
## med_fam_inc 1.317e-04 3.407e-04 0.386 0.69918
## pct_lt15k 6.081e-03 4.728e-03 1.286 0.19841
## num_prom 4.057e-03 2.562e-03 1.584 0.11327
## lifetime_gifts -2.889e-04 4.018e-04 -0.719 0.47218
## largest_gift 2.091e-03 3.342e-03 0.626 0.53152
## last_gift -1.399e-02 8.670e-03 -1.614 0.10656
## months_since_donate -5.405e-02 1.123e-02 -4.815 1.48e-06 ***
## time_lag 9.081e-05 7.709e-03 0.012 0.99060
## avg_gift -4.180e-03 1.234e-02 -0.339 0.73486
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 3327.0 on 2399 degrees of freedom
## Residual deviance: 3256.7 on 2385 degrees of freedom
## AIC: 3286.7
##
## Number of Fisher Scoring iterations: 4fund.glm1.probs = predict(fund.glm1, newdata = test20, type = "response")
fund.glm1.pred = rep("No Donor", length(fund.glm1.probs))
fund.glm1.pred[fund.glm1.probs > 0.5] = "Donor"
table(fund.glm1.pred, fund.test)## fund.test
## fund.glm1.pred Donor No Donor
## Donor 173 160
## No Donor 117 150(150 + 173)/600## [1] 0.5383333overall fraction of correct predictions from our first logistic regression model is 0.5383333.
fund.glm2 = glm(target_bin ~ num_child + income + months_since_donate, data = train80, family = binomial)
summary(fund.glm2)##
## Call:
## glm(formula = target_bin ~ num_child + income + months_since_donate,
## family = binomial, data = train80)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.6884 -1.1729 0.7957 1.1538 1.7361
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 2.24287 0.36427 6.157 7.41e-10 ***
## num_child -0.34048 0.12598 -2.703 0.00688 **
## income 0.05419 0.02508 2.160 0.03074 *
## months_since_donate -0.06656 0.01052 -6.329 2.47e-10 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 3327.0 on 2399 degrees of freedom
## Residual deviance: 3275.8 on 2396 degrees of freedom
## AIC: 3283.8
##
## Number of Fisher Scoring iterations: 4fund.glm2.probs = predict(fund.glm2, newdata = test20, type = "response")
fund.glm2.pred = rep("No Donor", length(fund.glm2.probs))
fund.glm2.pred[fund.glm2.probs > 0.5] = "Donor"
table(fund.glm2.pred, fund.test)## fund.test
## fund.glm2.pred Donor No Donor
## Donor 185 163
## No Donor 105 147(147 + 185)/600## [1] 0.5533333overall fraction of correction predictions from our second logistic regression model is 0.5533333.
library(class)
set.seed(12345)
fund.train.X = cbind(fundNew$num_child, fundNew$income, fundNew$months_since_donate)[train,]
fund.test.X = cbind(fundNew$num_child, fundNew$income, fundNew$months_since_donate)[-train,]
train.target = fundNew$target[train]
knn.pred1 = knn(fund.train.X, fund.test.X, train.target, k = 1)
table(knn.pred1, fund.test)## fund.test
## knn.pred1 Donor No Donor
## Donor 177 163
## No Donor 113 147(177 + 147)/600## [1] 0.54overall fraction of correct predictions is 0.54. The KNN model, with K = 1, correctly predicts the target variable 54% of the time
set.seed(12345)
knn.pred10 = knn(fund.train.X, fund.test.X, train.target, k = 10)
table(knn.pred10, fund.test)## fund.test
## knn.pred10 Donor No Donor
## Donor 184 159
## No Donor 106 151(184 + 151)/600## [1] 0.5583333mean(knn.pred10 != fund.test)## [1] 0.4416667overall fraction of correct predictions is 0.5583333. The KNN model, with K = 10, correctly predicts the target variable 55.8% of the time
set.seed(12345)
knn.pred100 = knn(fund.train.X, fund.test.X, train.target, k = 100)
table(knn.pred100, fund.test)## fund.test
## knn.pred100 Donor No Donor
## Donor 205 183
## No Donor 85 127(206 + 128)/600## [1] 0.5566667mean(knn.pred100 != fund.test)## [1] 0.4466667overall fraction of correct predictions is 0.5566667 The KNN model, with K = 100, correctly predicts the target variable 55.67% of the time
now use our best selected model to predict the outcome of whether the observations in future_fundraising.rds will be a donor or non-donor.
future_fundraising = readRDS("future_fundraising.rds")summary(future_fundraising)## zipconvert2 zipconvert3 zipconvert4 zipconvert5 homeowner num_child
## No :99 Yes:27 No :94 No :74 Yes:91 Min. :1.000
## Yes:21 No :93 Yes:26 Yes:46 No :29 1st Qu.:1.000
## Median :1.000
## Mean :1.067
## 3rd Qu.:1.000
## Max. :4.000
## income female wealth home_value med_fam_inc
## Min. :1.000 Yes:70 Min. :0.000 Min. : 0.0 Min. : 0.0
## 1st Qu.:2.750 No :50 1st Qu.:5.750 1st Qu.: 579.5 1st Qu.:276.2
## Median :4.000 Median :8.000 Median : 897.5 Median :359.0
## Mean :3.767 Mean :6.567 Mean :1093.7 Mean :384.7
## 3rd Qu.:5.000 3rd Qu.:8.000 3rd Qu.:1319.5 3rd Qu.:474.8
## Max. :7.000 Max. :9.000 Max. :5413.0 Max. :823.0
## avg_fam_inc pct_lt15k num_prom lifetime_gifts
## Min. : 0.0 Min. : 0.00 Min. : 15.00 Min. : 15.0
## 1st Qu.:311.8 1st Qu.: 5.00 1st Qu.: 29.75 1st Qu.: 45.0
## Median :405.5 Median :12.50 Median : 47.00 Median : 82.0
## Mean :426.5 Mean :14.52 Mean : 47.85 Mean :101.9
## 3rd Qu.:522.2 3rd Qu.:21.25 3rd Qu.: 64.00 3rd Qu.:126.5
## Max. :813.0 Max. :63.00 Max. :103.00 Max. :431.0
## largest_gift last_gift months_since_donate time_lag
## Min. : 5.00 Min. : 1.00 Min. :17.00 Min. : 0.000
## 1st Qu.: 10.00 1st Qu.: 7.00 1st Qu.:29.00 1st Qu.: 3.750
## Median : 15.00 Median : 12.00 Median :31.00 Median : 5.000
## Mean : 16.79 Mean : 14.52 Mean :31.24 Mean : 6.508
## 3rd Qu.: 20.00 3rd Qu.: 16.00 3rd Qu.:35.00 3rd Qu.: 9.000
## Max. :100.00 Max. :100.00 Max. :37.00 Max. :26.000
## avg_gift
## Min. : 2.963
## 1st Qu.: 6.819
## Median : 9.331
## Mean :11.235
## 3rd Qu.:13.354
## Max. :42.500library(class)
set.seed(12345)futureTrain = sample(120, 120)
future.test.X = cbind(future_fundraising$num_child, future_fundraising$income, future_fundraising$months_since_donate)[futureTrain,]
futureDonors = knn(fund.train.X, future.test.X, train.target, k = 10)
futureDonors_value = as.character(futureDonors)
futureDonors_value## [1] "No Donor" "No Donor" "Donor" "Donor" "No Donor" "No Donor"
## [7] "No Donor" "Donor" "Donor" "Donor" "Donor" "No Donor"
## [13] "Donor" "Donor" "Donor" "Donor" "No Donor" "No Donor"
## [19] "No Donor" "Donor" "Donor" "Donor" "No Donor" "Donor"
## [25] "Donor" "Donor" "Donor" "No Donor" "Donor" "No Donor"
## [31] "Donor" "Donor" "Donor" "No Donor" "Donor" "Donor"
## [37] "Donor" "No Donor" "Donor" "No Donor" "Donor" "No Donor"
## [43] "Donor" "Donor" "Donor" "Donor" "No Donor" "No Donor"
## [49] "Donor" "Donor" "Donor" "Donor" "No Donor" "No Donor"
## [55] "No Donor" "Donor" "Donor" "No Donor" "No Donor" "No Donor"
## [61] "Donor" "No Donor" "No Donor" "No Donor" "Donor" "Donor"
## [67] "No Donor" "Donor" "No Donor" "Donor" "Donor" "Donor"
## [73] "Donor" "Donor" "No Donor" "No Donor" "Donor" "No Donor"
## [79] "Donor" "No Donor" "No Donor" "Donor" "Donor" "No Donor"
## [85] "No Donor" "No Donor" "No Donor" "Donor" "Donor" "No Donor"
## [91] "Donor" "No Donor" "Donor" "No Donor" "Donor" "Donor"
## [97] "No Donor" "Donor" "No Donor" "Donor" "No Donor" "No Donor"
## [103] "No Donor" "Donor" "No Donor" "No Donor" "Donor" "Donor"
## [109] "No Donor" "No Donor" "Donor" "No Donor" "Donor" "No Donor"
## [115] "Donor" "Donor" "No Donor" "No Donor" "No Donor" "No Donor"write.table(futureDonors_value, file = "model_comp.csv", col.names = c("value"), row.names = FALSE)n = 3000 p = 20 y = target; this is a factor variable with two levels - Donor and No Donor
using our training dataset from prior, and future_fundraising.rds as our test dataset, our final model correctly predicted 54.17% of the response variable for 35% of the test data.
knn as the classification model worked best when attempting to predict a Donor or Non-donor. Most important factors are income, children, and those who havae donated in the past few months. Advertising directed towards these groups of people, like parenting magzines and children/ family friendly TV networks would work best.